UNIVERSITY OF PITTSBURGH
JJarlington JVleinorial J_(ibrary
LIBRARY
OF
Aboriginal American
Literature.
No. V.
EDITED BY
D. G. BRINTON, M. D.
PHILADELPHIA
1885.
BRINTON S LIBRARY OF
ABORIGINAL AMERICAN LITERATURE.
NUMBER V.
THE LENAPE
AND THEIR
LEGENDS;
WITH THE COMPLETE TEXT AND SYMBOLS
OF THE
WALAM GLUM,
A NEW TRANSLATION, AND AN INQUIRY INTO ITS AUTHENTICITY.
BY
DANIEL G. BRINTON, A.M., M.D.,
Professor of Ethnology and Archeology at the Academy of Natural
Sciences, Philadelphia.
President of the Numismatic and Antiquarian Society of Philadelphia ; Member of the
American Philosophical Society, the American Antiquarian Society, the Pennsyl-
vania Historical Society, etc.; Membre de la Societe Royale des Antiquaires du
Nord; Delegue General de I'lnstitution Ethnographique ; Vice-President du
Congres International des Americanistes ; Corresponding Member of the Anthro-
pological Society of Washington, etc.
D. G. BRINTON.
PHILADELPHIA.
1885.
-" 7 F 9
^^^9
&
Entered according to Act of Congress, in the year 1885, by
D. G, BRINTON,
In the Office of the Librarian of Congress. All rights reserved.
PREFACE.
In the present volume I have grouped a series of ethno-
logical studies of the Indians of Eastern Pennsylvania, New
Jersey and Maryland, around what is asserted to be one of
the most curious records of ancient American history.
For a long time this record — the Walam Olum, or Red
Score — was supposed to have been lost. Having obtained
the original text complete about a year ago, I printed a few
copies and sent them to several educated native Delawares
with a request for aid in its translation and opinions on its
authenticity. The results will be found in the following
pages.
The interest in the subject thus excited prompted me to a
general review of our knowledge of the Lenape or Dela-
wares, their history and traditions, their language and cus-
toms. This disclosed the existence of a number of MSS. not
mentioned in bibliographies, some in the first rank of im-
portance, especially in the field of linguistics. Of these I
have made free use.
In the course of these studies I have received suggestions
and assistance from a number of obliging friends, among
whom I would mention the native Delawares, the Rev. Albert
Anthony, and the Rev. John Kilbuck ; Mr. Horatio Hale
and the Right Rev. E. de Schweinitz ; Dr. J. Hammond
Trumbull, Prof. A. M. Elliott and Gen. John Mason Brown.
Not without hesitation do I send forth this volume to the
V
VI PREFACE.
learned world. Regarded as an authentic memorial, the
original text of the Walam Olum will require a n:iore accu-
rate rendering than I have been able to give it ; while the
possibility that a more searching criticism will demonstrate
it to have been a fabrication may condemn as labor lost the
pains that I have bestowed upon it. Yet even in the latter
case my work will not have been in vain. There is, I trust,
sufificient in the volume to justify its appearance, apart from
the Red Score ; and the latter, by means of this complete
presentation, can now be assigned its true position in Ameri-
can archaeology, whatever that may be.
CONTENTS.
CHAPTER I.— I r. The Algonkin Stock 9
Scheme of its Dialects. — Probable Primitive Location.
§ 2. The Iroquois Stock 13
The Susquehannocks. — The Hurons. — The Cherokees.
CHAPTER n. — The Wapanachki or Eastern Algonkin Con-
federacy 19
The Confederated Tribes. — The Mohegans.— The Nanticokes. — The Co-
noys. — The Shawnees. — The Saponies. — The Assiwikalees.
CHAPTER HI.— The Lenape or Delawares 33
Derivation of the Name Lenape. — The Three Sub-Tribes : the Minsi or
Wolf, the Unami or Turtle, and the Unalachtgo or Turkey Tribes. —
Their Totems. — The New Jersey Tribes: the Wapings, Sanhicans and
Mantas. — Political Constitution of the Lenape. — Vegetable Food Re-
sources. — Domestic Architecture. — Manufactures. — Paints and Dyes. —
Dogs. — Interments. — Computation of Time. — Picture Writing. — Record
Sticks. — Moral and Mental Character. — Religious Belief — Doctrine of
the Soul. — The Native Priests. — Religious Ceremonies.
CHAPTER IV. — The Literature and Language of the
Lenape 74
§ 1. Literature of the Lenape Tongue. — Campanius; Penn ; Thomas;
Zeisberger; Heckewelder; Roth; Ettwein ; Grube; Dencke;
Luckenbach ; Henry ; Vocabularies ; a Native Letter.
g 2. General Remarks on the Lenape.
g 3. Dialects of the Lenape.
g 4. Special Structure of the Lenape.— The Root and the Theme ;
Prefixes; Suffixes; Derivatives; Grammatical Notes.
CHAPTER v.— Historical Sketches of the Lenape 109
g I. The Lenape as "Women."
g 2. Recent Migrations of the Lenape.
g 3. Missionary Efforts in the Provinces of Pennsylvania and New
Jersey.
vii
Vlll CONTENTS.
PAGE
CHAPTER VI.— Myths and Traditions of the Lenape 130
Cosmogonical and Culture Myths.— The Culture-hero, Michabo.— Myths
from Lindstrom, Ettwein, Jasper Donkers, Zeisberger. — Native
Symbolism. — The Satumian Age. — Mohegan Cosmogony and Mi-
gration Myth.
National Traditions. — Beatty's Account. — The Number Seven. — Hecke-
welder's Account. — Prehistoric Migrations. — Shavi^nee Legend. —
Lenape Legend of the Naked Bear.
CHAPTER VII.— The Walam Olum : Its Origin, Authen-
ticity AND Contents 148
Biographical Sketch of Rafinesque. — Value of his Writings.— His ac-
count of the Walum Olum. — Was it a Forgery ? — Rafinesque's
Character. — The Text Pronounced Genuine by Native Delawares.
— Conclusion Reached.
Phonetic System of the Walum Olum. — Metrical Form. — Pictographic
System. — Derivation and Precise Meaning of Walum Olum. — The
MS. of the Walam Olum. — General Synopsis of the Walam
Olum. — Synopsis of its Parts.
THE WALUM OLUM.— Original Text and Translation 169
Notes 219
Vocabulary 233
Appendix 255
Index 257
THE LENAPE
AND THEIR LEGENDS.
CHAPTER I.
§ I. The Algonkin Stock.
Scheme of its Dialects. — Probable Primitive Location.
§ 2. The Iroquois Stock.
The Susquehannocks. — The Hurons. — The Cherokees.
§ I. The Algonkin Stock.
About the period 1 500-1 600, those related tribes whom we
now know by the name of Algonkins were at the height of
their prosperity. They occupied the Atlantic coast from the
Savannah river on the south to the strait of Belle Isle on the
north. The whole of Newfoundland was in their possession ;
in Labrador they were neighbors to the Eskimos ; their north-
ernmost branch, the Crees, dwelt along the southern shores
of Hudson Bay, and followed the streams which flow into it
from the west, until they met the Chipeways, closely akin
to themselves, who roamed over the water shed of Lake Su-
perior. -The Blackfeet carried a remote dialect of their tongue
quite to the Rocky Mountains ; while the fertile prairies of
Illinois and Indiana were the homes of the Miamis. The
area of Ohio and Kentucky was very thinly peopled by a few
B 9
10 THE LENAPE AND THEIR LEGENDS.
of their roving bands; but east of the Alleghanies, in the
valleys of the Delaware, the Potomac and the Hudson, over
the barren hills of New England and Nova Scotia, and
throughout the swamps and forests of Virginia and the Caro-
linas, their osier cabins and palisadoed strongholds, their
maize fields and workshops of stone implements, were num-
erously located.
It is needless for my purpose to enumerate the many small
tribes which made up this great group. The more prominent
were the Micmacs of Nova Scotia, the Abnakis of Maine, the
Pequots and Narragansets, in New England, the Mohegans
of the Hudson, the Lenape on the Delaware, the Nanticokes
around Chesapeake Bay, the Pascataway on the Potomac,
and the Powhatans and Shawnees further south; while
between the Great Lakes and the Ohio river were the
Ottawas, the Illinois, the Pottawatomies, the Kikapoos,
Piankishaws, etc.
The dialects of all these were related, and evidently at
some distant day had been derived from the same primitive
tongue. Which of them had preserved the ancient forms
most closely, it may be premature to decide positively, but
the tendency of modern studies has been to assign that place
to the Cree — the northernmost of all.
We cannot erect a genealogical tree of these dialects. It
is not probable that they branched off, one after another,
from a common stock. The ancient tribes each took their
several ways from a common centre, and formed nuclei for
subsequent development. We may, however, group them in
such a manner as roughly to indicate their relationship. This
I do on the following page : —
THE ALGONKIN STOCK. H
Cree,
Old Algonkin,
Montagnais.
'^ Chipeway,
Ottawa,
Pottawattomie,
Miami,
Peoria,
Pea,
• Piankishaw,
'. Kaskaskia,
Menominee,
Sac,
Fox,
Kikapoo.
Sheshatapoosh,
Secoffee,
Micmac,
Melisceet,
Etchemin,
Abnaki.
Mohegan,
Massachusetts,
Shawnee,
Minsi, ]
Unami,
Unalachtigo, j
Nanticoke,
Powhatan,
Pampticoke,
Blackfoot,
Gros Ventre,
Shevenne,
12 THE LENAPE AND THEIR LEGENDS.
Granting, as we must, some common geographical centre
for these many dialects, the question where this was located
becomes an interesting one.
More than one attempt to answer it has been made. Mr.
Lewis H. Morgan thought there was evidence to show that
the valley of the Columbia river, Oregon, "was the initial
point from which the Algonkin stock emigrated to the great
lake region and thence to the Atlantic coast. "^ This is in
direct conflict with the evidence of language, as the Blackfoot
or Satsika is the most corrupt and altered of the Algonkin
dialects. Basing his argument on this evidence, Mr. Horatio
Hale reaches a conclusion precisely the reverse of that of
Morgan. " The course of migration of the Indian tribes,"
writes Mr. Hale, " has been from the Atlantic coast westward
and southward. The traditions of the Algonkins seem to
point to Hudson's Bay and the coast of Labrador."'^ This
latter view is certainly that which accords best with the testi-
mony of language and of history.
We know that both Chipeways and Crees have been
steadily pressing westward since their country was first
explored, driving before them the Blackfeet and Dakotas.^
The Cree language is built up on a few simple, unchange-
able radicals and elementary words, denoting being, relation,
energy, etc. ; it has extreme regularity of construction, a
1 Lewis H. Morgan, Indian Migrations, in Beach's Indian Miscellany,
p. 218.
2 H. Hale, Indian Migrations as Evidenced by Language, p. 24.
(Chicago, 1883.)
3 See the R. P. A. Lacombe Dictionnaire de la Langue des Cris.
Introd., p. xi. (Montreal, 1874.)
THE SUSQUEHANNOCKS. 13
single negative, is almost wholly verbal and markedly incor-
porative, has its grammatical elements better defined than its
neighbors, and a more consistent phonetic system.^ For
these and similar reasons we are justified in considering it the
nearest representative we possess of the pristine Algonkin
tongue, and unless strong grounds to the contrary are
advanced, it is proper to assume that the purest dialect is
found nearest the primeval home of the stock.
§ 2. The Iroquois Stock.
Surrounded on all sides by the Algonkins were the Iroquois,
once called the Five or Six Nations. When first discovered
they were on the St. Lawrence, near Montreal, and in the
Lake Region of Central New York. Various other tribes,
not in their confederacy, and generally at war with them,
spoke dialects of the same language. Such were the Hurons
or Wyandots, between the Georgian Bay and Lake Erie, the
Neutral Nation on the Niagara river, the Fries on the
southern shore of the lake of that name, the Nottoways in
Virginia, and the Tuscaroras in North Carolina. The
Cherokees, found by the whites in East Tennessee, but
whose national legend, carefully preserved for generations,
located them originally on the head waters of the Ohio, were
a remote offshoot of this same stem.
The Susquehannocks.
The valley of the Susquehanna river was occupied by a
tribe of Iroquois lineage and language, known as the Susque-
1 See Joseph Howse, A Grammar of the Cree Language, p. 13, et al.
(London, 1842.)
14 THE LENAPE AND THEIR LEGENDS.
hannocks, Co?iestogas and Andastes. The last name is .Iro-
quois, from andasta, a cabin pole. By some, " Susquehan-
nock" has also been explained as an Iroquois word, but its
form is certainly Algonkin. The terminal k is the place-
sign, hanna denotes a flowing stream, while the adjectival
prefix has been identified by Heckewelder with schachage,
straight, from the direct course of the river near its mouth,
and by Mr. Guss with woski, new, which, he thinks, referred
to fresh or spring water.
Of these the former will appear the preferable, if we allow
for the softening of the gutturals, which was a phonetic trait
of the Unami dialect of the Lenape.
The Susquehannocks were always at deadly feud with the
Iroquois, and between wars, the smallpox and the whites,
they were finally exterminated. The particulars of their
short and sad history have been presented with his character-
istic thoroughness by Dr. John G. Shea,^ and later by Prof.
N. L. Guss.^ They were usually called by the Delawares
Mengwe, which was the term they applied to all the Iroquois-
speaking tribes.' The English corrupted it to Minqua and
1 In a note to Mr. Gowan's edition of George Alsop's Province of
Maryland, pp. 117-121 (New York, 1869); also, in 1858, in an article
" On the Identity of the Andastas, Minquas, Susquehannocks, and Con-
estogas," in the Amer. Hist. Mag., Vol. II, p. 294.
- Early Indian History on the Susquehanna, p. 31. (Harrisburg,
1S83.)
3 Mengwe is the Onondaga yetikwe, males, or men, viri, and was
borrowed from that dialect by the Delawares, as a general term. Bishop
Ettwein states that the Iroquois called the Delawares, Mohegans, and all
the New England Indians Agozhagduta.
THE SUSQUEHANNOCKS. 15
Mingo, and as the eastern trail of the Susquehannocks lay
up the Conestoga Creek, and down the Christina, both
those streams were called "Mingo Creek" by the earlv
settlers.
It is important for the ethnology of Pennsylvania, to under-
stand that at the time of the first settlement the whole of
the Susquehanna Valley, from the Chesapeake to the New
York lakes, was owned and controlled by Iroquois-speaking
tribes. A different and erroneous opinion was expressed
by Heckewelder, and has been generally received. He
speaks of the Lenape Minsi as occupying the head waters
of the Susquehanna. This was not so in the historic
period.
The claims of the Susquehannocks extended down the
Chesapeake Bay on the east shore, as far as the Choptank
River, and on the west shore as far as the Patuxent. In
1654 they ceded to the government of Maryland their
southern territory to these boundaries. ^ The first English
explorers met them on the Potomac, about the Falls, and the
Pascatoways were deserting their villages and fleeing before
them, when, in 1634, Calvert founded his colony at St.
Mary's.
Their subjection to the Five Nations took place about 1680,
and it was through the rights obtained by this conquest that,
at the treaty of Lancaster, 1 744, Canassatego, the Onondaga
speaker for the Nation, claimed pay from the government of
Maryland for the lands on the Potomac, or, as that river was
called in his tongue, the Cohojtgorontas.
1 Bozman, History of Maryland, Vol. I, p. 167.
16 THE LENAPE AND THEIR LEGENDS.
The Hurons.
The Hurons, Wyandots, or Wendats, were another Iroquois
people, who seem, at some remote epoch, to have come into
contact with the Lenape. The latter called them Delamat-
tenos^ and claimed to have driven them out of a portion of
their possessions. A Chipeway tradition also states that the
Hurons were driven north from the lake shores by Algonkin
tribes.^ We know, from the early accounts of the Jesuits,
that there was commercial intercourse between them and the
tribes south of the lakes, the materials of trade being princi-
pally fish and corn.'' The Jesuit Relations of 1648 contain
quite a full account of a Huron convert who, in that year,
visited the Lenape on the Delaware River, and had an inter-
view with the Swedish Governor, whom he took to task for
neglecting the morals of his men.
The Cherokees.
The Cherokees were called by the Delawares Kittuwa
{Kuttoowauw, in the spelling of the native Aupaumut).
This word I suppose to be derived from the prefix, kit, great,
and the root tawa (Cree, yette, tawa), to open, whence
tawatawik, an open, /. e., uninhabited place, a wilderness
(Zeisberger).
The designation is geographical. According to the tradi-
tion of the Cherokees, they once lived (probably about the
1 Heckewelder, History of the Indian Nations, p. 80.
2 Peter Jones, History of the Ojibway Nation, p. 32.
3 Relation des Jesuites, 1637, p. 154. The Hurons, at that time, are
stated to have had reliable traditions running back more than two hundred
years. Relation de 1639, p. 50.
THE CHEROKEES. 17
fourteenth century) in the Ohio Valley, and claimed to have
been the constructors of the Grave Creek and other earth-
works there. ^ Some support is given to this claim by the
recent linguistic investigations of Mr. Horatio Hale,^ and
the archaeological researches of Prof. Cyrus Thomas.' They
were driven southward by their warlike neighbors, locating
their council fire first near Monticello, Va., and the main
body reaching East Tennessee about the close of the fif-
teenth century. As late as 1 730 some of them continued to
live east of the Alleghanies, while, on the other hand, it is
evident, from the proper names preserved by the chroniclers
of De Soto's expedition (1542), that at that period others
held the mountains of Northern Georgia. To the Delawares
they remained kit-tawa-wi, inhabitants of the great wilder-
ness of Southern Ohio and Kentucky,
Delaware traditions distinctly recalled the period when
portions of the Cherokees were on the Ohio, and recounted
1 " The Cherokees had an oration, in which was contained the history
of their migrations, which was lengthy." This tradition related " that they
came from the upper part of the Ohio, where they erected the mounds on
Grave Creek, and that they removed hither [to East Tennessee] from the
country where Monticello is situated." This memory of their migra-
tions was preserved and handed down by official orators, who re-
peated it annually, in public, at the national festival of the green corn
dance. J. Haywood, Natural and Aboriginal History of Tennessee, pp.
224-237. (Nashville, 1823.) Haywood adds: "It is now nearly
forgotten." I have made vain attempts to recover some fragments of i
from the present residents of the Cherokee Nation.
2 Indian Migrations as Evidenced by Language, p. 22.
3 Prof. Thomas has shown beyond reasonable doubt that the Cherokees
were mound builders within the historic period.
18 THE LENAPE AND THEIR LEGENDS.
long wars with them.^ When the Lenape assumed the office
of peacemaker, this feud ceased,, and was not renewed until
the general turmoil of the French-Indian wars, 1 750-60. After
this closed, in 1768, the Cherokees sought and effected a re-
newal of their peaceful relations with the Delawares, and in
1779 they even sent a deputation of "condolence" to their
"grandfather," the Lenape, on the death of the head chief.
White Eyes.^
^ Loskiel, Geschichte der Mission, etc., p. 160; Heckewelder, ^zj^?^
of the Indian Nations, p. 54. Bishop Ettwein states that the last Chero-
kees were driven from the upper Ohio river about 1700-10. His essay on
the " Traditions and Languages of the Indian Nations," written for General
Washington, in 1788, was first published in the Bulletin of the Pa. Hist.
Soc, 1844.
2 Heckewelder, Indian N'ations, pp. 88, 327. Mr. H. Hale, in The
Iroquois Book of Rites, has fully explained the meaning and importance
of the custom of "condolence." The Stockbridge Indian, Aupaumut, in
his Jozirnal, writes of the Delawares, that when they lose a relative, " ac-
cording to ancient custom, long as they are not comforted, they are not to
speak in public, and this ceremonie of comforting each other is highly
esteemed among these nations." Xarrative of Hetidrick Aupauniiit, in
Mems. Hist. Soc. Pa., Vol. II, p. 99.
CHAPTER II.
The Wapanachki or Eastern Algonkin Confederacy.
The Confederated Tribes. — The Mohegans. — The Nanticokes. — The Conoys. — The
Shawnees. — The Saponies. — The Assiwikalees.
The Confederated Tribes.
All the Algonkin nations who dwelt north of the Potomac,
on the east shore of Chesapeake Bay, and in the basins of the
Delaware and Hudson rivers, claimed near kinship and an
identical origin, and were at times united into a loose, defens-
ive confederacy.
By the western and southern tribes they were collectively
known as Wapanachkik — "those of the eastern region" —
which in the form Abnaki is now confined to the remnant of
a tribe in Maine. The Delawares in the far West retain tra-
ditionally the ancient confederate name, and still speak of
themselves as "Eastlanders" — 0-puh-narke. (Morgan.)
The members of the confederacy were the Mohegans
(Mahicanni) of the Hudson, who occupied the valley of that
river to the falls above the site of Albany, the various New
Jersey tribes, the Delawares proper on the Delaware river and
its branches, including the Minsi or Monseys, among the
mountains, the Nanticokes, between Chesapeake Bay and the
Atlantic, and the small tribe called Canai, Kanawhas or Gan-
awese, whose towns were on tributaries of the Potomac and
Patuxent.
That all these were united in some sort of an alliance, with
19
20 THE LENAPE AND THEIR LEGENDS.
the Delawares at its head, is not only proved by the traditions
of this tribe itself, but by the distinct assertion of the Mohe-
gans and others, and by events within historical times, as the
reunion of the Nanticokes, New Jersey and Eastern Indians
with the Delawares as with the parent stem.^
The Mohegans.
The Mohegans, Mo-he- kun-ne-uk, dwelt on the tide-waters
of the Hudson, and from this their name was derived. Dr.
Trumbull, indeed, following Schoolcraft, thinks that they
" took their tribal name from maingan, a wolf, and Mohe-
ganick r= Chip, maniganikan, ' country of wolves. ' " ^ They,
themselves, however, translate it, "seaside people," or more
fully, " people of the great waters which are constantly ebbing
* Heckewelder, History of the Indian Nations, p. 60, and Narrative
of Hendrick Aupaumut, 1 79 1, in Mems. Hist. Soc. Fa., Vol. II. The
latter, himself a native Mohegan, repeatedly refers to " the ancient cove-
nant of our ancestors," by which this confederacy was instituted, which
included the " Wenaumeew (Unami), the Wemintheew (Minsi), the
Wenuhtokowuk (Nanticokes) and Kuhnauwantheew (Kanawha)." From
-old Pennsylvania documents, Proud gives the members of the confederacy
or league as " the Chiholacki or Delawares, the Wanami, the Munsi, the
Mohicans and Wappingers." History of Penna., Vol. II, p. 297, note.
Compare J. Long, Voyages and Travels, p. 10 (London, 1791), who gives
the same list. Mr. Ruttenber writes : " In considering the political rela-
tions of the Lenapes, they should be considered as the most formidable of
the Indian confederacies at the time of the discovery of America, and as
having maintained for many years the position which subsequently fell
to the Iroquois." — Indian Tribes on Hudson River, p. 64.
• Trumbull, Indian Names in Connecticut, p. 31. Schoolcraft had
already given the same derivation in his History and Statistics of the
Indian Tribes.
THE MOHEGANS. 21
or flowing."! The compound is machaak, great, hickan,
tide ("ebbing tide," Zeis.; "tide of flood," Campanius)
and ik, animate plural termination.
The Mohegans on the Hudson are said to have been
divided into three phratries, the Bear, the Wolf and the
Turtle, of whom the Bear had the primacy.^ Mr. Morgan,
however, who examined, in i860, the representatives of the
nation in Kansas,^ discovered that they ,had precisely the
same phratries as the Delawares, that is the Wolf, the Turtle,
and the Turkey, each subdivided into three or four gentes.
He justly observes that this " proves their immediate connec-
tion with the Delawares and Munsees by descent," and thus
renders their myths and traditions of the more import in the
present study.
Linguistically, the Mohegans were more closely allied to
the tribes of New England than to those of the Delaware
Valley, Evidently, most of the tribes of Massachusetts and
Connecticut were comparatively recent offshoots of the parent
stem on the Hudson, supposing the course of migration had
been eastward.
In some of his unpublished notes Mr. Heckewelder identi-
fies the Wampanos, who lived in Connecticut, along the
shore of Long Island Sound, and whose council fire was
where New Haven now stands, as Mohegans, while the
Wapings or Opings of the Northern Jersey shore were a mixed
! Capt. Hendricks, in Mass. Hist. Soc. Colls., Vol. IX, p. loi. Lewis
H. Morgan, Systems of Consanguinity and Affinity, p. 289.
^ Ruttenber, History of the Indian Tribes of Hudson's River, p. 50.
^ Morgan, Ancient Society, pp. 173-4.
22 THE LENAPE AND THEIR LEGENDS.
clan derived from intermarriages between Mohegans and
Monseys.^
The Nanticokes.
The Nanticokes occupied the territory between Chesapeake
Bay and the ocean, except its southern extremity, which
appears to have been under the control of the Powhatan tribe
of Virginia.
The derivation of Nanticoke is from the Delaware Unechtgo,
'' tide- water people," and is merely another form of Una-
lachtgo, the name of one of the Lenape sub-tribes. In both
cases it is a mere geographical term, and not a national eponym.
In the records of the treaty at Fort Johnston, 1757, the
Nanticokes are also named Tiawco. This is their Mohegan
name, Otaydchgo,vi\\\c\\ means "bridge people," or bridge
makers, the reference being to the skill with which the Nanti-
cokes could fasten floating logs together to construct a bridge
across a stream. In the Delaware dialect this was Tawach-
1 These opinions are from a MS. in the library of the American Philo-
sophical Society, in the handwriting of Mr. Heckewelder, entitled Notes,
Amendments and Additions to Heckewelder^ s History of the Indians (8vo,
pp. 38.) Unfortunately, this MS. was not placed in the hands of Mr.
Reichel when he prepared the second edition of Heckewelder' s work
for the Historical Society of Pennsylvania.
An unpublished and hitherto unknown work on the Mohegan language
is the Aliscellanea Lingiue Natio7iis Indices Alahikan dietce, curd siiscepta
a Joh. Jac. Schmick, 2 vols., small 8vo. ; MS. in the possession of the
American Philosophical Society. Schmick was a Moravian missionary,
born in 1714, died 1778. He acquired the Mohegan dialect among the
converts at Gnadenhiitten. His work is without date, but may be placed
at about 1765. It is grammatical rather than lexicographical, and offers
numerous verbal forms and familiar phrases.
THE NANTICOKES. 23
guano, from taiachquoan, a bridge. The latter enables us to
identify the Tockwhoghs, whom Captain John Smith met on
the Chesapeake, in 1608, with the Nanticokes. The Kus-
carawochs, whom he also visited, have been conclusively
shown by Mr. Bozman^ to have been also Nanticokes,
By ancient traditions, they looked up to the Lenape as their
"grandfather," and considered the Mohegans their "breth-
ren." ^ That is, they were, as occasion required, attached
to the same confederacy.
In manners and customs they differed little from their north-
ern relatives. The only peculiarity in this respect which
is noted of them was the extravagant consideration they be-
stowed on the bones of the dead. The corpse was buried for
some months, then exhumed and the bones carefully cleaned
and placed in an ossuary called man-to-kimip (== 77ianito, with
the locative termination, place of the mystery or spirit).
When they removed from one place to another these bones
were carried with them. Even those who migrated to northern
Pennsylvania, about the middle of the last century, piously
brought along these venerable relics, and finally interred them
near the present site ?f Towanda, whence its name, Tawun-
deunk, "where we bury our dead."*
1 J. ^ozm'a.u, History of Maryland, No\. I, pp. 1 12, 1 14, 121, 177.
This laborious writer still remains the best authority on the aboriginal
inhabitants of Maryland.
2 " The We nuh tok o wuk are our brothers according to ancient agree-
ment." Journal of Hendrick Atipan??mf, Mcms. Hist. Soc. Pa., Vol. II,
P- 77-
3 Charies Beatty, Journal of a Journey, etc., p. 87. Heckewelder,
Indian Nations, pp. 90, et seq. Ibid. Tra7is. Am. Phil. Soc, Vol. IV, p. 362.
24 THE LENAPE AND THEIR LEGENDS.
Their dialect varied considerably from the Delaware, of
, which it is clearly a deteriorated form. It is characterized by
abbreviated words and strongly expirated accents, as tah /
quah / quah ! su, short ; quah / nah ! gut, long.
Our knowledge of it is limited to a few vocabularies. The
earliest was taken down by Captain John Smith, during his
exploration of the Chesapeake. The most valuable is one
obtained by Mr. William Vans Murray, in 1792, from the
remnant in Maryland. It is in the library of the American
Philosophical Society, and has never been correctly or com-
pletely printed.
The Nanticokes broke up early. Between the steady en-
croachments of the whites and the attacks of the Iroquois
they found themselves between the upper and the nether
millstones.
According to their own statement to Governor Evans, at a
conference in 1707, they had at that time been tributary to
the latter for twenty-seven years, /. <?., since 1680. Their last
head chief, or "crowned king," Winicaco, died about 1720.
A few years after this occurrence bands of them began to re-
move to Pennsylvania, and at the middle of the century
were living at the mouth of the Juniata, under the immediate
control of the Iroquois. Thence they removed to Wyoming,
and in 1753, "in a fleet of twenty-five canoes," to the Iro-
quois lands in western New York. Others of their nation
were brought there by the Iroquois in 1767; but by the
close of the century only five families survived in that
region.^
1 The authorities for these facts are Bozman, History of Maryland,
Vol. I, pp. 175-180; Heckewelder, Indian Nations, pp. 93, sqq. ; E. de
THE CONOYS. 25
A small band called the Wiwash remained on Goose creek,
Dorchester county, Maryland, to the same date.
The Coftoys.
The fourth member of the Wapanachki was that nation
variously called in the old records Conoys, Ganawese or
Canaways, the proper form of which Mr. Heckewelder states
to be Canai}
Considerable obscurity has rested on the early location and
affiliation of this people. Mr. Heckewelder vaguely places
them '' at a distance on the Potomac," and supposes them to
have been the Kanawhas of West Virginia.'^ This is a loose
guess. They were, in fact, none other than the Piscataways
of Southern Maryland, who occupied the area between
Chesapeake Bay and the lower Potomac, about St. Mary's, and
along the Piscataway creek and Patuxent river.
Proof of this is furnished by the speech of their venerable
head chief, "Old Sack," at a conference in Philadelphia in
1743.^ His words were : ''Our forefathers came from Pis-
Schweinitz, Life of Zeisberger, pp. 208, 322, etc.; the Treaty Records,
and MSS. in the library of the American Philosophical Society.
That the Nanticokes came from the South into Maryland has been
maintained, on the ground that as late as 1770 they claimed land in North
Carolina. New York Colonial Documents, Vol. VIII, p. 243. But the
term "Carolina " was, I think, used erroneously in the document referred
to, instead of Maryland, where at that date there were still many of the
tribe.
1 History of the Indian Nations, Introduction, p. xlii.
* Ibid., pp. 90-122.
» Minutes of the Provincial Council of Penna., Vol. IV, p. 657.
C
26 THE LENAPE AND THEIR LEGENDS.
catua to an island in Potowmeck ; and from thence down to
Philadelphia, in old Proprietor Penn's time, to show their
friendship to the Proprietor. After their return they brought
down all their brothers from Potowmeck to Conejoholo, on
the east side Sasquehannah, and built a town there."
This interesting identification shows that they were the
people whom Captain John Smith found (1608) in numerous
villages along the Patuxent and the left bank of the lower
Potomac. The local names show them to have been of
Algonkin stock and akin to the Nanticokes.
Conoy, Ganawese, Kanawha, are all various spellings of a
derivative from an Algonkin root, meaning " it is long"
(Del. guncu, long, Cree kinowaw, it is long,) and is found
applied to various streams in Algonkin territory.^
Piscataway, or Pascatoway, as it is spelled in the early
narratives, also recurs as a local name in various parts of the
Northern States. It is from the root pashk, which means to
separate, to divide. Many derivatives from it are in use in
the Delaware tongue. In the Cree we have the impersonal
form, pakestikweyaw, or the active animate pasketiwa, in the
sense of " the division or branch of a river. "^ The site of
Further proof of this in a Treaty of Peace concluded in 1682 by the New
York colonial government, between the Senecas and Maryland Indians.
In this instrument we find this tribe referred to as " the Canowes alias
Piscatowayes," and elsewhere as the " Piscatoway of Cachnawayes."
New York Colonial Documents, Vol. Ill, pp. 322, 323.
1 I am aware that Mr. Johnston, deriving his information from Shawnee
interpreters, translated the name Kanawha, as " having whirlpools."
iTrans. of the Amer. Antiq. Soc, Vol. I, p. 297.) But I prefer the
derivation given in the text.
* Lacombe, Dictionnaire de la Langue des Cris, s. v. In Delaware the
THE CONOYS. 27
Kittamaquindi {kittamaque-hik, Great Beaver Place,) the so-
called "metropolis of Pascatoe,"^ was where Tinker's creek
and Piscataway creek branch off from their common estuary,
about fifteen miles south of Washington city.
The ''emperor" Chitomachen, Strong Bear (chtiani, strong,
macha, bear), who bore the title Tayac (Nanticoke, tallak,
head chief) ruled over a dominion which extended about 130
miles from east to west.
The district was thinly peopled. On the upper shores of
the west side of the Chesapeake Captain John Smith and the
other early explorers found scarcely any inhabitants. In
1 63 1 Captain Henry Fleet estimated the total number of
natives "in Potomack and places adjacent," at not over
5000 persons.^ This included both sides of the river as high
up as the Falls, and the shores of Chesapeake Bay.
Chitomachen, with his family, was converted to the Catholic
faith in 1640, by the exertions of the Jesuit missionary, Father
Andrew White, but died the year after. When the English
first settled at St. Mary's, the tribe was deserting its ancient
seats, through fear of the Susquehannocks, and diminished
rapidly after that date.
root takes the form pack, from which are derived, by suffixes, the words
pach-at,ios^\it,pachgeechen,v^\^trQ the road branches oii,packshican,a.
knife — something that divides, etc.
1 Relatio Itineris in Marylandiam, p. 63. (Edition of the Md. Hist.
Soc. 1874.)
2 See \^x^ Journal, published in NeiU's Fotmders of 3faryland {Mh^ny,
1876). Fleet was a prisoner among the Pascatoways for five years, and
served as an interpreter to Calvert's colony.
28 THE LENAPE AND THEIR LEGENDS.
Father White was among them from 1634 to 1642, and
composed a grammar, dictionary and catechism of their
tongue. Of these, the catechism is yet preserved in manu-
script, in the library of the Domus Professa of the Jesuits, in
Rome. It would be a great benefit to students of Algonkin
dialects to have his linguistic works sought out and published.
How far his knowledge of the language extended is uncertain.
In a letter from one of the missionaries, dated 1642, who
speaks of White, the writer adds: "The difficulty of the
language is so great that none of us can yet converse with the
Indians without an interpreter." ^
That it was an Algonkin dialect, closely akin to the Nanti-
coke, is clear from the words and proper names preserved in
the early records and locally to this day. The only word
which has created doubts has been the name of "a certain
imaginary spirit called Ochre ^ ^ It has been supposed that
this was the Huron oki. But it is pure Algonkin. It is the
Cree oki-sikow {etre dtt del, ange, Lacombe), the Abnaki
ooskoo {katini ooskoo, Bon Esprit, matsini ooskoo, Mauvais
Esprit, Rasles).
It was nearly allied to that spoken in Virginia among Pow-
1 Relaiio Itineris in Marylandiam, p. 84. The Rev. Mr. Kampman,
at one time Moravian missionary among the Delawares, told me that even
with the modern aids of grammars, dictionaries and educated native
instructors, it is considered to require five years to obtain a sufficient know-
ledge of their language to preach in it. The slowness of the early Mary-
land priests to master its intricacies, therefore, need not surprise us.
2 " Omni vero ratione placare conantur phantasticum quemdam spiritum
quem Ochre nominant, ut ne noceat." Relatio Itineris in Marylandiam,
p. 40.
THE SHAWNEES. 29
hatan's subjects, as an English boy who had lived with that
chieftain served as an interpreter between the settlers and the
Patuxent and neighboring Indians.^
The Conoys were removed, before 1743, from Conejoholo to
Conoy town, further up the Susquehanna, and in 1 744 they
joined several other fragmentary bands at Shamokin (where
Sunbury, Pa. , now stands). Later, they became merged with
the Nanticokes.^
The Shawnees.
The wanderings of the unstable and migratory Shawnees
have occupied the attention of several writers, but it cannot
be said that either their history or their affiliations have been
satisfactorily worked out.^
Their dialect is more akin to the Mohegan than to the
Delaware, and when, in 1692, they first appeared in the area
of the Eastern Algonkin Confederacy, they came as the
friends and relatives of the former.*
They were divided into four bands, as follows : —
1. Piqua, Y>'<^o'^ex\y Fikoweu, "he comes from the ashes."
2. Mequachake, "a. fat man filled," signifying completion
or perfection. This band held the privilege of the hereditary
priesthood.
^ Bozman, History of Maryland, Vol. I, p. 166.
2 " The Nanticokes and Conoys are now one nation." Minutes of the
Provincial Council of Fenna., 1759, Vol. VIII, p. 176.
3 On this tribe see " The Shawnees and Their Migrations," by Dr.
D. G. Brinton, in the American Historical Magazine, 1866; M. F. Force,
Some Early Notices of the Indians of Ohio. Cincinnati, 1S79.
* See Colonial History of New York, Vol. IV. Index. Loskiel,
Geschichte der Mission, etc., p. 25.
30 THE LENAPE AND THEIR LEGENDS.
3. Kiscapocoke.
4. Chilicothe.^
Of these, that which settled in Pennsylvania was the
Pikoweu, who occupied and gave their name to the Pequa
valley in Lancaster county.^
According to ancient Mohegan tradition, the New England
Pequods were members of this band. These moved east-
wardly from the Hudson river, and extended their conquests
over the greater part of the area of Connecticut. Dr.
Trumbull, however,^ assigns a different meaning to their
name, and a more appropriate one — Pequttdog, the De-
stroyers. Some countenance is given to the tradition by the
similarity of the Shawnee to the Mohegan, standing, as it
does, more closely related to it than to the Unami Delaware.
It has been argued that a band of the Shawnees lived in
Southern New Jersey when that territory first came to the
knowledge of the whites. On a Dutch map, drawn in 1614
or thereabouts, a tribe called Saw wanew is located on the left
1 These names are as given by John Johnston, Indian agent, in 1819.
Archaologia Americana, Vol. I, p. 275. Heckewelder says they had
four divisions, but mentions only two, the Pecuwesi and Woketamosi.
(MSS. in Lib. Am. Philos. Soc.)
2 " That branch of Shawanos which had settled part in Pennsylvania
and part in New England were of the tribe of Shawanos then and ever
since called Pi'co7veu or Pe'koweu, and after emigrating to the westward
settled on and near the Scioto river, where, to this day, the extensive flats
go under the name of ' Pickoway Plains.' " Heckewelder MSS. in Lib.
Am. Phil. Soc.
s In a note to Roger Williams, Key into the Language of America, p.
22. The tradition referred to is mentioned in the Heckewelder MSS.
THE SAPOONES. 31
bank of the Delaware river, near the Bay;^ and DeLaet
speaks of the Sawanoos as living there.
I am inclined to believe that, in both these cases, the term
was used by the natives around New York Bay in its simple
geographical sense of ''south" or " southern," and not as a
tribal designation. It frequently appears with this original
meaning in the Walam Olum.
The Sapoonees.
A tribe called the Sapoonees, or Saponies, is mentioned as
living in Pennsylvania, attached to the Delawares, about the
middle of the last century.^
They are no doubt the Saponas who once dwelt on a branch
of the Great Pedee river in North Carolina, and who moved
north about the year 1720.^ They were said to have joined
the Tuscaroras, but the Pennsylvania records class them with
the Delawares. Others, impressed by the similarity of
'^di-po-nees to Pa-nis, have imagined they were the Pawnees,
now of the west. There is not the slightest importance to be
attached to this casual similarity of names.
They were called, by the Iroquois, Tadirighrones, and were
distinctly identified by them with the nation known to the
English as the Catawbas.* For a long time the two nations
carried on a bitter warfare.
1 Printed in the Colonial History of New York, Vol. I. Compare
Force, ubi supra, pp. 16, 17.
2 Rev. J. Morse, Report on Indian Affairs, p. 362.
5 See Gallatin, Synopsis of the Indian Tribes, pp. 85, 86.
* See New York Colonial Documents, Vol. V, pp. 660, 673, etc.
32 THE LENAPE AND THEIR LEGENDS.
The Assiwikales.
This band of about fifty families, or one hundred men
(about three hundred souls), are stated to have come from
South Carolina to the Potomac late in the seventeenth century,
and in 1731 were settled partly on the Susquehanna and partly
on the upper Ohio or Alleghany. Their chief was named
Aqueioma, or Achequeloma.
Their name appears to be a compound of assin, stone, and
wikwam, house, and they were probably Algonkin neighbors
of the Shawnees in their southern homes, and united with
them in their northern migration.^
^ Pennsylvania Archives, Vol. I, pp. 299, 300, 302. Gov. Gordon
writes to the "Chiefs of ye Shawanese and Assekelaes," under date De-
cember, 1 73 1, "I find by our Records that about 34 Years since some
Numbers of your Nation came to Sasquehannah," etc. Ibid., p. 302.
CHAPTER III.
The Lenape or Delawares.
Derivation of the Name Lenape. — The Three Sub-Tribes : the Minsi or Wolf, the
Unami or Turtle, and the Unalachtgo or Turkey Tribes. — Their Totems. —
The New Jersey Tribes : the Wapings, Sanhicans and Manias. — Political Con-
stitution of the Lenape. — Vegetable Food Resources. — Domestic Architecture. —
Manufactures. — Paints and Dyes. — Dogs. — Interments. — Computation of
Time. — Picture Writing. — Record Sticks.— Moral and Mental Character. —
Religious Belief. — Doctrine of the Soul. — The Native Priests. — Religious
Ceremonies.
Derivation of Lenni Lenape.
The proper name of the Delaware Indians was and is
Lenape {a as in father, ^ as ^ in mate). Dr. J. Hammond
Trumbull ^ is quite wide of the mark both in calling this a
''misnomer," and in attributing its introduction to Mr.
Heckewelder.
Long before that worthy missionary was born, the name
was in use in the official documents of the commonweahh of
Pennsylvania as the synonym in the native tongue for the
Delaware Indians,' and it is still retained by their remnant
1 See his remarks in the Transactions of the Americafi Philological
Association, 1872, p. 157.
2 For instance, in Governor Patrick Gordon's Letter to the Friends,
1728, where he speaks of " Our Lenappys or Delaware Indians," in
Fenna. Archives, Vol. I, p. 230. At the treaty of Easton, 1756, Tedyus-
>ung, head chief of the Delawares, is stated to have represented the
"Lenopi" Indians {Minutes of the Council, Phila., 1757), and in the
"Conference of Eleven Nations living West of Allegheny," held at
Philadelphia, 1759, the Delawares are included under the tribal name
33
34 THE LENAPE AND THEIR LEGENDS.
in Kansas as the proper term to designate their collective
nation, embracing its sub-tribes.^
The derivation of Lenape has been discussed with no little
learning, as well as the adjective lennt, which often precedes
it (Lenni Lenape). Mr. Heckewelder stated that lemii means
"original, pure," and that Lenape signifies "people."^ Dr.
Trumbull, in the course of a long examination of the words
for "man" in the Algonkin dialects, reaches the conclusion
that "Len-ape" denotes "a common adult male," /. e., an
Indian man ; letino lenape, an Indian of our tribe or nation,
and, consequently, vir, "a man of men. "^ He derives these
two words from the roots len (= nen^, a pronominal possessive,
and ape, an inseparable generic particle, " denoting an adult
male."
1 differ, with hesitation, from such an eminent authority ;
but this explanation does not, to my mind, give the precise
meaning of the term. No doubt, both lentio, which in Dela-
ware means man, and len, in Lenape, are from the pronomi-
nal radicle of the first person ne, I, we, mine, our. As the
native considered his tribe the oldest, as well as the most
important of created beings, "ours" with him came to be
synonymous with what was esteemed ancient, indigenous,
primeval, as well as human, man-like, /ar excellence. "We"
"Leonopy." See Minutes of the Provincial Council of Pen7ta.,Vo\.
VIII, p. 418.
^ So Mr. Lewis H. Morgan says, and he obtained the facts on the spot.
" Len-a'-pe was their former name, and is still used." Systems of Con-
sanguinity and Affinity, p. 289 (Washington, 187 1 ).
2 History of the Indian Nations, p. 401 .
^ Transactions of the American Philological Association, 1 871, p. 144,
DERIVATION OF LENNI LENAPE. 35
and ''men" were to him the same. The initial / is but a
slight modification of the n sound, and is given by Campanius
as an r, "rhenus, homo."
Lenape, therefore, does not mean "a common adult male,"
but rather " a male of our kind," or " our men."^
The termination ape is said by Heckewelder to convey
the idea of "walking or being in an erect posture." A
comparison of the various Algonkin dialects indicates that
it was originally a locative, signifying staying in a place,
abiding or sitting. Thus, in Cree, apu, he is there ; in
Chipeway, abi, he is at home; in Delaware, n" dappin, I am
here. The transfer of this idea to the male sex is seen in
the Cree, ap, to sit upon, to place oneself on top, apa, to
cover (animate and active); Chipeway, nabe, the male of
quadrupeds, Baraga says that for a Chipeway woman to
call her husband nin nabem (lit. ray coverer, comp. French,
femme couverie), is coarse.
1 Zeisberger's translation of Lenni Lenape as "people of the same
nation," would be more literal if it were put "men of our nation."
President Stiles, in his Itinerary, makes the statement: " The Delaware
tribe is called Poh-he-gan or Mo-hee-gan by themselves, and Aitquitsau-
ion." I have not been able to reach a satisfactory solution of the first
and third of these names.
That the Delawares did use the term Lenape as their own designation,
is shown by the refrain of one of their chants, preserved by Heckewelder.
It was —
"Hiisca n' lenape-win"
Truly I — a Lenape — am.
Or: "I am a true man of our people." Trans. Amer. Philos. Soc,
Vol. IV, N. Ser., p. 381.
36 THE LENAPE AND THEIR LEGENDS.
The Lenape Sub- Tribes.
The Lenape were divided into three sub-tribes: —
1. The Minsi, Monseys, Montheys, Munsees, or Minisinks.
2. The Unami, or Wonameys.
3. The Unalachtigo.
No explanation of these designations will be found in Hecke-
welder or the older writers. From investigations among living
Delawares, carried out at my request by Mr. Horatio Hale,
it is evident that they are wholly geographical, and refer to
the locations of these sub-tribes on the Delaware river.
Mi?isi, properly Minsiu, and formerly Minassiniu, means
"people of the stony country," or briefly, "mountaineers."
It is a synthesis of minthiu, to be scattered, and achsin, stone,
according to the best living native authorities.^
Undmi, or Wndmiu, means "people down the river,"
from naheu, down-stream.
Unalachtigo, properly W^naldchtko, means "people who
live near the ocean," from wimalawat, to go towards, and
t^kow or fkbUy wave.
Historically, such were the positions of these sub-tribes
when they first came to the knowledge of Europeans.
The Minsi lived in the mountainous region at the head
waters of the Delaware, above the Forks, or junction of the
Lehigh river. One of their principal fires was on the
Minisink plains, above the Water Gap, and another on the
1 Mr. Eager, in his History of Orange County, quotes the old surveyor,
Nicolas Scull (1730), in favor of translating minisink "the water is
gone; " and Ruttenber, in his Histojy of the Native Tribes of the Hudson
River, supposes that it is derived from menatey^ an island. Neither of
these commends itself to modern Delawares.
THE LENAPE SUB-TRIBES. 37
East Branch of the Delaware, which they called Namaes Sipu,
Fish River. Their hunting grounds embraced lands now in
the three colonies of Pennsylvania, New York and New
Jersey. The last mentioned extinguished their title in 1758,
by the payment of one thousand pounds.
That, at any time, as Heckewelder asserts, their territory
extended up the Hudson as far as tide-water, and westward
" far beyond the Susquehannah," is surely incorrect. Only
after the beginning of the eighteenth century, when they
had been long subject to the Iroquois, have we any
historic evidence that they had a settlement on the last
named river.
The Unamis' territory on the right bank of the Delaware
river extended from the Lehigh valley southward. It was
with them and their southern neighbors, the Unalachtigos,
that Penn dealt for the land ceded him in the Indian Deed
of 1682. The Minsis did not take part in the transaction,
and it was not until 1737 that the Colonial authorities
treated directly with the latter for the cession of their
territory.^
The Unalachtigo or Turkey totem had its principal seat on
the affluents of the Delaware near where Wilmington now
stands. About this point. Captain John Smith, on his map
(1609,) locates the Chikahokin. In later writers this name is
spelled Chihohockies, Chiholacki and Chikolacki, and is stated
by the historians Proud and Smith to be synonymous with
Delawares.^ The correct form is Chikelakt, from chik'eno,
^ See Penna. Archives, Vol. I, pp. ^\o-l.
2 Proud, History of Penna., Vol. II, p. 297 ; S. Smith, Hist, of New
Jersey, p. 456 ; Henry, Diet, of the Delaware Lang., MS., p. 539.
38 THE LENAPE AND THEIR LEGENDS,
turkey, the modern form as given by Whipple/ and aki
land. The n, I and r were alternating letters in this
dialect.
The population was, however, very sparse, owing to the
predatory incursions of the Susquehannocks, whose trails,
leading up the Octorara and Conestoga, and down the Chris-
tina and Brandywine Creeks, were followed by war parties
annually, and desolated the west shores of the Bay and lower
river. When, in 1634, Captain Thomas Young explored the
river, the few natives he found on the west side told him
(through the medium of his Algonkin Virginian interpreter)
that the "Minquaos" had killed their people, burnt their
villages, and destroyed their crops, so that " the Indians had
wholly left that side of the river which was next their
enemies, and had retired themselves on the other side farre
up into the woods. "^
North of the Chikelaki, Smith's map locates the Macocks.
This name does not appear in later authors, but near that site
were the Okahoki band, who occupied the shores of Rid4ey
and Crum creeks and the land between them. There they
remained until 1703, when they were removed to a small
reservation of 500 acres in what is now Willistown township,
Chester county.^
1 Delaware Vocabulary in Whipple, Ewbank & Turner's Report, 1855.
The German form is tsichemim.
* A Brief Relation of the Voyage of Captayne Thomas Yong, in Mass.
Hist. Soc. Colls. , 4th series, Vol. IX, p. 119.
* See the original Warrant of Survey and Minutes relating thereto, in
Dr. George Smith's History of Delaware County, Pa., pp. 209, 210
(Phila., 1862). The derivation is uncertain. Captain John Smith gives
THE TOTEMIC ANIMALS. 39
The Totemic Anitnals.
These three sub-tribes had each its totemic animal, from
which it claimed a mystical descent. The Minsi had the
Wolf, the Unami the Turtle, and the Unalachtigo the Turkey.
The Unamis claimed and were conceded the precedence of
the others, because their ancestor, the Turtle, was not the
common animal, so-called, but the great original tortoise
which bears the world on its back, and was the first of living
beings, as I shall explain on a later page.
In referring to the totemic animals the common names
were not used, but metaphorical expressions. Thus the Wolf
was referred to as Ptuksit, Round Foot {ptuk, round, sit, foot,
from the shape of its paws ;) the turtle was Fakoango, the
Crawler ; and the turkey was Pullaeu, he does not chew,^
referring to the bird's manner of swallowing food.
The signs of these animals were employed in their picture
writing, painted on their houses or inscribed on rocks, to
designate the respective sub-tribes. But only in the case of
the Unamis was the whole animal represented. The Turkey
mahcawq for pumpkin, and this appears to be the word in the native name
of Chester Creek, Macopanackhan, which is also seen in Marctis Hook.
(See Smith's Hisi. Del. Co., pp. 145, 381.) I am indined to identify the
Macocks with the M'okahoki as "the people of the pumpkin place," or
where those vegetables were cultivated.
1 The Shawnee word is the same, pellewaa, whence their name for the
Ohio River, Pellewaa seepee, Turkey River. (Rev. David Jones, Jourtial of
Two Visits Made to Some Nations of hidians on ike West Side of the River
Ohio in 1772 and 1773, p. 20.) From this is derived the shortened form
Plaen, seen in Playwickey, or Pla7iwikit, the town of those of the Turkey
Tribe, in Berks county, Pa. (Heckewelder, Indian Names, p. 355.)
40 THE LENAPE AND THEIR LEGENDS.
tribe painted only one foot of their totemic bird, and the
Minsi the extended foot of the wolf, though they sometimes
added an outline of the rest of the animal/
These three divisions of the Lenape were neither " gentes"
nor "phratries," though Mr. Morgan has endeavored to
force them into his system by stating that they were " of the
nature of phratries."^ Each was divided into twelve
families bearing female names, and hence probably referring
to some unexplained matriarchal system. They were, as I
have called them, sub-tribes. In their own orations they
referred to each other as '' playmates." (Heckewelder.)
The New Jersey Lenape.
The native name of New Jersey is given as Sha'akbee
(English orthography : a as in fate) ; or as the German mis-
sionaries wrote it, Sche'jachbi. It is a compound of bi, water,
aki, land, and the adjective prefix schey, whi.ch means some-
thing long and narrow {scheyek, a string of wampum ; schaje-
li?iquaii, the edge of the eyes, the eyelids, etc.) This would
be equivalent to "long-land water," and, according to the
rules of Delaware grammar, which place the noun used in
the genitive sense before the noun which governs it, the term
would be more suitable to some body of water, Delaware bay
or the ocean, than to the main land.
The Lenape distinctly claimed the whole of the present
area of New Jersey, Their great chief, Tedyuscung, stated
at the Conference at Easton (1757), that their lands reached
eastward to the shore of the sea. The New Jersey tribes
1 Heckewelder, Hist. Indian Nations, pp. 253-4.
2 Lewis H. Morgan, Ancient Society, pp. 1 7 1-2.
THE NEW JERSEY LENAPE. 41
fully recognized their unity. As early as 1694, at an inter-
view with Governor Markham at Philadelphia, when the
famous Taraany and other Lenape chieftains were present,
Mohocksey, a chief of the Jersey Indians, said: "Though
we live on the other side of the water (/. e., the Delaware
river), yet we reckon ourselves all one, because," he added,
giving a characteristically native reason, ''because we drink
one water. "^
The names, number and position of the Jersey tribes have
not been very clearly made out.
A pamphlet published in London, in 1648, states that there
were twenty-three Indian kinglets in its area, with about 2000
warriors in all. Of these. Master Robert Evelin, a surveyor,
who spent several years in the Province about 1635, names
nine on the left bank of the Delaware, between Cape May
and the Falls. The names are extremely corrupt, but it may
be worth while giving them.^
1. Kechemeches, 500 men, five miles above Cape May.
2. Manteses, 100 bowmen, twelve leagues above the
former.
3. Sikonesses.
4. Asomoches, 100 men.
5. Eriwoneck, 40 men.
6. Ramcock, 100 men.
1 Minutes of the Provincial Council of Pennsylvania, July 6th, 1694.
2 Master Evelin's Letter is printed in Smith's History of New Jersey,
2d ed. Some doubt has been cast on his letter, because of its connection
with the mythical " New Albion," but his personality and presence on
the river have been vindicated. See The American Historical Magazine,
Vol. I, 2d series, pp. 75, 76.
D
42 THE LENAPE AND THEIR LEGENDS.
7. Axion, 200 men.
8. Calcefar, 150 men.
9. Mosilian, 200 men, at the Falls.
Of these, the Mantes lived on Salem creek ; Ramcock is
Rancocas creek ; the Eriwoneck are evidently the Ermomex
of Van der Donck's map of 1656; Axion may be for Assis-
cunk creek, above Burlington, from Del. assiscu, mud ; assis-
cunk, a muddy place. Lindstrom and Van der Donck name
the most Southern tribe in New Jersey Naraticons. They
were on and near Raccoon creek, which on Lindstrom' s map
is Narratlcon Sipit, the Naraticon river. Probably the Eng-
lish name is simply a translation of the Del. nachemim, rac-
coon.
In 1675 the number of sachems in Jersey of sufficient im-
portance for the then Governor Andros to treat with were
four. It is noted that when he had made them the presents
customary on such occasions, "They return thanks and fall a
kintacoying, singing keno?i, kcnon.'''' ^ This was the Delaware
genan {genama, thank ye him. Zeis).
The total number in New Jersey a few years before this
(1671) were estimated by the authorities at " about a thousand
persons, besides women and children." ^
The " JVapi//gs, Opings or Fompfons,^^ as they are named
in the old records, were the tribe which dwelt on the west
shore of New York harbor and southwardly, or, more
exactly, "from Roeloff Jansen's Kill to the sea."^ They
were of the Minsi totem, and were the earliest of the
1 Nezv Jersey Archives, Vol. I, p. 183.
2 Ibid, Vol. I, p. 73.
* Ruttenber, Hist, of the Indian Ti'ibes of Hudson Riner. s. v.
THE NEW JERSEY LENAPE. 43
Lenape who saw white men, when, in 1524, the keel of
Verrazano was the first to plough the waters of New York
harbor.
The name Waping or Oping is derived from Wapan, east,
and was applied to them as the easternmost of the Lenape
nation.^ Their other name, Porapton, Mr. Heckewelder
identifies with pih?n-tom, crooked-mouthed, though its appli-
cability is not obvious.^
In the middle of the eighteenth century the remains of the
Pompton Indians resided on the Raritan river. The bound-
aries of their territory were defined in 1756, at the Treaty of
Crosswicks.
The Sajihicans occupied the Delaware shore at the Falls,
near where Trenton now stands, and extended eastward along
the upper Indian path quite to New York bay. Heckewelder
says that this name, SanJzhicani, means a gun lock, and was
applied by the Lenape to the Mohawks who were first furnished
with muskets by the Europeans. This has led some writers
to locate a band of Mohawks at the Falls.
The Sanhicans were, however, undoubtedly Lenape. Cam-
panius, who quotes the name of the place in 1642, classes
1 Heckewelder, in his unpublished MSS,, asserts that both these names
mean "Opossum." It is true that the name of this animal in Lenape is
woapink, in the New Jersey dialect opiing, and in the Nanticoke of Smith
oposon ; but all these are derived from the root w«i5, which originally
meant "white," and was applied to the East as the place of the dawn
and the light. The reference is to the light gray, or whitish, color of the
animal's hair. Compare the Cree, wapiskowes, cendre, il a le poll blafard.
Lacombe, Dictionnaire de la Langue des Oris. s. v.
2 On Indian Names, p. 375, in Trans. American Philosophical Society,
Vol. IH, n. ser.
44 THE LENAPE AND THEIR LEGENDS.
them as such. In Van der Donck's map, of 1656, they are
marked as possessing the land at the Falls and Manhattan
Bay; and De Laet gives the numerals and a number of words
from their dialect, which are all pure Delaware, as : —
Sanhican.
Delaware.
Deer,
atto,
achtu.
Bear,
machquoyuo.
machquak.
Wolf,
metumnu,
raetemmeu.
Turkey,
sickenum.
tschickenum,
Their name has lost its first syllable. It should be assan-
hican. This means not merely and not originally a gun-
flint, but any stone implement, from achsin, or, in the New
Jersey dialect, assim, a stone, and hican, an instrument.
They were distinctively " the stone-implement people."
This is plainly with reference to their manufactures near
Trenton. The great deposit of post-glacial gravels at this
point abound with quartzite fragments suitable for working
into stone implements, and to what extent they were utilized
by the natives is shown by the enormous collection, num-
bering over thirty thousand specimens, which Dr. Charles C.
Abbott, of Trenton, has made in that immediate vicinity. A
horde of over 125 beautifully chipped lance heads of quartz
and jasper, and the remains of a workshop of remarkable
magnitude, were evidences of the extensive manufacture that
once prevailed there.
The left bank of the Delaware, from the vicinity of Bur-
lington quite to and below Salem, was held by a warlike tribe
known to the settlers as the Manias, or Ma?ifos, or Afandcs,
otherwise named the Frog Indians. They extended eastward
THE NEW JERSEY LENAPE. 45
along the main or southern Indian path, which led from the
Delaware, below the mouth of Rancocas Creek, to the
extensive Indian plantations or corn fields near Sandy Hook,
mentioned by Campanius and Lindstrom/
Mr. Henry has derived their name from mangi, great,^ and
others have suggested menatey, an island; but I do not
think either of these is tenable. I have no doubt that mante
is simply a mis-spelling of monthee, which is the form given
by the East Jersey and Stockbridge Indians to the name of
the Minsi or Monsey sub-tribe of the Delawares.^ This is
further indicated by the fact that toward the beginning
of the eighteenth century they incorporated themselves
wholly with the two other Lenape sub-tribes.* We thus
find that the Minsis were not confined to the North and
Northwest, as Heckewelder and others wrote, but had pressed
southward in New Jersey, quite to the shores of Delaware
Bay.
The New Jersey Indians disappeared rapidly. As early as
1 721 an official document states that they were "but few,
and very innocent and friendly."' When, in 1745, the
missionary Brainerd visited their settlement at Crosweeksung,
Burlington county, he found some '' who had lived with the
1 Proud, History of Pennsylvania, Vol. 1. 144, H, P- 295- Heckewelder,
Trans. Am. Philo. Soc.,^o\. IV, p. 376.
2 Matthew G. Henry, Delaivare Indian Dictionary, p. 709- (MS. in
the Library of the Am. Phil. Soc.)
3 "The Monthees who we called Wemintheuw," etc. Journal of
Hendrick Aupawniit, Meins. Hist. Soc. Pa., Vol. H.p. 77-
4 Heckewelder, ubi stiprd.
5 New Jersey Archives, Vol. V, p. 22.
46 THE LENAPE AND THEIR LEGENDS.
white people under gospel light, had learned to read, were
civil, etc."^ Those with whom he labored at this place
subsequently removed to New Stockbridge, Mass., and united
with the Mohegans and others there. ^
The Swedish traveler, Peter Kalm, who spent about a year
in New Jersey in 1749, observes that the disappearance of
the native population was principally due to two agencies.
Smallpox destroyed "incredible numbers," "but brandy
has killed most of the Indians."^
The dialect of the New Jersey Indians was soft and vocalic,
avoiding the gutturals of their northern relatives, and with-
out the frequent unpleasant forcible expirations of the Nanti-
coke. A vocabulary of it, obtained for Mr. Thomas Jefferson,
in 1792, at the village of Edgpiiliik, West New Jersey, is in
MS. in the library of the American Philosophical Society.
Political Cojistitution.
Each totem of the Lenape recognized a chieftain, called
sachem, sakima, a word found in most Algonkin dialects,
with slight variations (Chip, ogima, Cree, okimaw, Pequot,
sachimmd), and derived from a root oki, signifying above in
space, and by a transfer frequent in all languages, above in
power. Thus, in Cree,* we have sdkamow, " il projecte, il
^ T/ie Rise and Progress of a Remarkable Work of Grace Among the
Indians. By David Brainerd, in Works, p. 304.
2 E. de Schweinitz, Life of Zeisberger, p. 660, note.
* Travels into North America, Vol. II, pp. 93-94 (London, 1771).
* Lacombe, Dictionnaire de la Langiie des Cris, p. 71 1. Dr. Trumbyll,
however, maintains that it is derived from sohkau-au, he prevails over
(note to Roger Williams' Key, p. 162). If there is a genetic connection,
POLITICAL CONSTITUTION. 47
montre la tete," and in Delaware, ui' ochgitschi, the part
above, the upper part (Zeisberger), etc.
It appears from Mr. Morgan's inquiries, that at present
and of later years, "the office of sachem is hereditary in
the gens, but elective among its members."^ Loskiel, how-
ever, writing on the excellent authority of Zeisberger, states
explicitly that the chief of each totem was selected and
inaugurated by those of the remaining two.^ By common
and ancient consent, the chief selected from the Turtle totem
was head chief of the whole Lenape nation.
These chieftains were the "peace chiefs." They could
neither go to war themselves, nor send nor receive the war
belt — the ominous string of dark wampum, which indicated
that the tempest of strife was to be let loose. Their proper
badge was the wampum belt, with a diamond-shaped figure
in the centre, worked in white beads, which was the symbol
of the peaceful council fire, and was called by that name.
War was declared by the people at the instigation of the
" war captains," valorous braves of any birth or family who
had distinguished themselves by personal prowess, and espe-
cially by good success in forays against the enemy.^
the latter is the derivative. The word sakhna is not known among the
Minsi. In place of it they say k'Jitai, the great one, from kehtan, great.
From this comes the corrupted forms iayach or tallach of the Nanticokes,
and the tayac of the Pascatoways.
1 Lewis H. Morgan, Ancient Society, p. 172.
2 Loskiel, Geschichte der Mission, p. 168.
3 For these particulars see Ettwein, Traditions and Language of the
Indians^ in Bulletin of the Pa. Hist. Soc.,No\. I; Charles Y,tM.y , Journal
of a Tour, etc., p. S'-
48 THE LENAPE AND THEIR LEGENDS.
Nor did the authority of the chiefs extend to any infringe-
ment on the traditional rights of the gens, as, for instance,
that of blood revenge. The ignorance of this limitation of
the central power led to various misunderstandings at the
time, on the part of the colonial authorities, and since then,
by later historians. Thus, in 1728, "the Delaware Indians
on Brandywine" were summoned by the Governor to answer
about a murder. Their chief, Civility, answered that it was
committed by the Minisinks, "over whom they had no
authority."^ This did not mean but that in some matters
authority could be exerted, but not in a question relating to
a feud of blood.
Agriculture and Food Resources.
The Lenape did not depend solely on the chase for sub-
sistence. They were largely agricultural, and raised a variety
of edible plants. Indian corn was, as usual, the staple; but
in addition to that, they had extensive fields of squashes,
beans and sweet potatoes.^ The hardy variety of tobacco
was also freely cultivated.
The value of Indian corn, the Zea mats, must have been
known to the Algonkin tribes while they still formed one
nation, as the same name is applied to it by tribes geo-
graphically the widest apart. Thus the Micmacs of Nova
Scotia call it pe-ds'kumun-ul whose theme ds'ku-inun re-
^ C. Thompson, Inquiry into the Causes of the Alienation of the Dela-
ware and Shawnee Indians, p. 16.
2 I assign them the sweet potato on the excellent authority of Dr. C.
Thompson, Essay on Indian Affairs, in Colls, of the Hist. Sac. of Penna.,
Vol. I, p. 81.
AGRICULTURE AND FOOD RESOURCES. 49
appears in the wuskaniiem (Elliott) and the scannemeneash
(Roger Williams) of New England, in the Delaware jesquem
(Campanius), and chasquem (Zeis.), and even in the Piegan
Blackfoot esko-tope.
The first radical a^/^. Chip, ashk, Del. aski, means ''green."
The application is to the green waving plant, so conspicuous
in the fields during the summer months. The second mim or
min is a generic suffix applied to all sorts of small edible fruits.
In the Blackfoot its place is supplied by another, and in the
Unami Delaware it is abbreviated to the letter in.
On the other hand, in the Chipeway word for corn, man-
damin, Ottawa mindamin, Cree mattamin, the second radical
is retained in full, while for the first is substituted an abbre-
viation of manito, divine ("it is divine, supernatural, or
mysterious"); if we may accept the opinion of Mr. School-
craft, and I know of no more plausible etymology.
Tobacco was called by the Delawares ksclia-tey, Zeis.,
scka-ta. Camp., or in the English orthography j-Z^z/fa;/!? (Vocab.
N. J. Inds.), and koshdhtahe (Cummings). I am inclined to
think that these are but dialectic variations and different
orthographies of the root Ua or ''dam {a nasal) found in
the New England witttdm-anog, Micmac tuinawa, Abnaki
wK daman (Rasle), Crtt tchistemaw, Chip, assema (= aste-
maw), Bla.ckfoot J>i-sid-kan ; a root which Dr. J. H. Trumbull
has satisfactorily identified as meaning "to drink," the smoke
being swallowed and likened to water. "To drink tobacco "
was the usual old English expression for " to smoke."
If this etymology is correct, it leads to the inference that
tobacco also was known to the ancient Algonkins before they
split up into the many nations which we now know, and
50 THE LENAPE AND THEIR LEGENDS.
furthermore that they must have lived in a region where these
two semi-tropical or wholly tropical plants, Indian corn and
tobacco, had been already introduced and cultivated by some
more ancient race. To conclude that they themselves brought
them from a tropical land, would be too hazardous.
The pipes in which the tobacco was smoked were called
appooke (modern Delaware o^pahokuii, Cumings' Vocab.)
They were of earthenware and of stone ; sometimes, it is said,
of copper. According to Kalm, the ceremonial pipes were
of a red stone, possibly the western pipe stone, and were very
highly prized.^
Of wild fruits and plants they consumed the esculent and
nutritious tubers on the roots of the Wild Bean, Apios iiiberosa,
the large, oval, fleshy roots of the arrow-leaved Sagittaria, the
former of which the Indians called Jwbbenis, and the latter
katniss, names which they subsequently applied to the
European turnip. They also roasted and ate the acrid
cormus of the Indian turnip. Arum triphylluni, in Delaware
taw-ho, taw-hin or tuck-ah, and collected for food the seeds
of the Golden Club, Oi'ontium aquaticiim, common in the
pools along the creeks and rivers. Its native name was
taw-kee?
House Building.
In their domestic architecture they differed noticeably
from the Iroquois and even the Mohegans. Their houses
were not communal, but each family had its separate resi-
dence, a wattled hut, with rounded top, thatched with mats
1 Peter Kahn, Travels in North America, Vol. II, p. 42.
2 See Peter Kalm, Travels in North America, Vol. II, pp. 110-I15 ;
William Darlington, Flora Cestrica. (West Chester, Pa., 1837.)
MANUFACTURES. 51
woven of the long leaves of the Indian corn or the stalks of
the sweet flag {A corns calamus,) or of the bark of trees
(anacon, a mat, Z.) These were built in groups and sur-
rounded with a palisade to protect the inhabitants from
sudden inroads.^
In the centre was sometimes erected a mound of earth,
both as a place of observation and as a location to place the
children and women. The remains of these circular ram-
parts enclosing a central mound were seen by the early
settlers at the Falls of the Delaware and up the Lehigh
valley.
Ma7iufactures.
The art of the potter was known and extensively practiced,
but did not indicate any unusual proficiency, either in the
process of manufacture or in the methods of decoration,
although the late Mr. F. Peale thought that, in the latter
respect, the Delaware pottery had some claims to a high
rank.^ The representation of animal forms was quite unusual,
only some few and inferior examples having been found.
^ For these facts, see Bishop Ettwein's article on the Traditions and
Languages of the Indians, Bulletin of the Pa. Hist. Soc, 1S48, p. 32.
Van der Donck (1656) describes these palisaded strongholds, and Cam-
panius (1642-4S) gives a picture of one. See also E. de Schvveinitz,
Life of Zeisberger, p. 83. The Mohegan houses were sometimes 180
feet long, by about 20 feet wide, and occupied by numerous families.
Van der Donck, Descrip. of the New Nether la7tds, pp. 196-7. Coll. N.
Y. Hist. Soc, Ser. II, Vol. I.
The native name of these wooden forts was 7ne7iachk, derived from
manachen, to cut wood (Cree, manikka, to cut with a hatchet). Roger
Williams calls them aumansk, a form of the same word.
* See the communication on " Pottery on the Delaware," by him, in the
52 THE LENAPE AND THEIR LEGENDS.
Their skill in manufacturing bead work and feather
mantles, and in dressing deer skins, excited the admiration
of the early voyagers. Although their weapons and utensils
were mostly of stone, there was a considerable supply of
native copper among them, in use as ornaments, for arrow
heads and pipes. Some specimens of it have been found by
Dr. Abbott near Trenton, and by other collectors in Penn-
sylvania,' and its scarcity in modern collections is to be
attributed to its being bought up and melted by the whites
rather than to its limited employment.
Soap stone was hollowed out with considerable skill, to
form bowls, and the wood of the sassafras tree was highly
esteemed for the same purpose (Kalm).
The maize was broken up in wooden or stone mortars with
a stone pestle, the native name of which was J>oco/iaac, a word
signifying also the virile member.
Proceedings of the Jm. Phil. Soc, 1868. The whole subject of the
archseology of the Delaware valley and New Jersey has been treated in
the most satisfactory manner by the distinguished antiquary, Dr. Charles
C. Abbott, in his work, Primitive Industry (Salem, Mass., 1881), and his
Stone Age in Neio Jersey (1877).
1 Four specimens are reported from Berks Co., Pa., by Prof. D. P.
Brunner, in his volume. The Indians of Bei-ks Co., Pa., pp. 94, 95
(Reading, 1881). The.se were an axe, a chisel, a knife and a gouge.
The metal was probably in part obtained in New Jersey, in part imported
from the Lake Superior region. See further, Abbott, Pri7nitive Industry,
chap, xxviii. Peter Kalm, the Swedish naturalist, who visited New Jer-
sey in 1748, says that when the copper mines " upon the second river
between Elizabeth Town and New York" were discovered, old mining
holes were found and tools which the Indians had made use of. Travels
in North America, Vol. I, p. 384.
PAINTS AND DYES. 53
Their arms were the war club or tomahawk, tomhickim, the
bow, hattape, and arrow, alluns, the spear, ta7iganaoim, and
for defence Bishop Ettwein states they carried a round shield
of thick, dried hide.
The spear was also used for spearing fish, which they,
moreover, knew how to catch with "brush nets," and with
fish hooks made of bone and the dried claws of birds
(Kalm).i
• Paints and Dyes.
The paints and dyes used by the Lenape and neighboring
Indians were derived both from the vegetable and mineral
realms. From the former they obtained red, white and blue
clays, which were in such extensive demand that the vicinity
of those streams in New Castle county, Delaware, which
are now called White Clay Creek and Red Clay Creek, was
widely known to the natives as Walamink, the Place of
Paint.
The vegetable world supplied a variety of dyes in the
colored juices of plants. These were mixed with the acid
juice of the wild, sweet-scented crab apple {Fyrus coronaria;
in Lenape, tombic'anall), to fix the dye.
A red was yielded by the root of the Sangiiinaria Canaden-
sis, still called "Indian paint root;" an orange by the root
of Phytolacca decandra, the poke or pocoon ; a yellow by the
1 Some antiquaries appear to have doubted whether the spear was in
use as a weapon of war among the Pennsylvania Indians (See Abbott,
Prhnitive Industry, p. 248.) But the Susquehannocks are distinctly
reported as employing as a weapon " a strong and light spear of locust
wood." Relatio Itineris in Marytandiam, p. 85.
54 THE LENAPE AND THEIR LEGENDS.
root oi Hydrastis Canadensis ; a black by a mixture of sumac
and white walnut bark, etc.^
Dogs.
The only domestic animal they possessed was a small
species of dogs with pointed ears. These were called aihim,
and were preserved less for protection or for use in hunting
than for food, and especially for ceremonial purposes.^
Interments.
The custom of common ossuaries for each gens appears to
have prevailed among the Lenape. Gabriel Thomas states
that ; " If a person of Note dies very far away from his place
of residence, they will convey his Bones home some consid-
erable Time after, to be buried there. "^ Bishop Ettwein
speaks of mounds for common burial, though he appears to
limit their use to times of war.*
^ For further information on this subject, an article may be consulted
in the Transactions of the American Philosophical Society, 1st Ser.,
Vol. in, pp. 222, et seq., by Mr. Hugh Martin, entitled " An Account of
the Principal Dies employed by the American Indians."
2 The Delawares had three words for dog. One was allum, which
recurs in many Algonkin dialects, and is derived by Mr. Trumbull from
a root signifying " to lay hold of," or " to hold fast." The second was
lennochum or Icnchiim, which means " the quadruped belonging to man;"
lenno, man; chum, a four-footed beast. The third was moekaneu, a
name derived from a general Algonkin root, in Cree, mokku, meaning
" to tear in pieces," from which the Delaware word for bear, inachque,
has its origin, and also, significantly enough, the verb "to eat" vet some
dialects.
3 History of West New Jersey, p. 3 (London, 169S).
* Bulletin Hist. Soc. of Penna., 1848, p. 32.
COMPUTATION OF TIME. 5^
One of these communal graveyards of the Minsis covers an
area of six acres on the Neversink creek/ while, according to
tradition, another of great antiquity and extent was located
on the islands in the Delaware river, above the Water-Gap.^
Computation of Time.
The accuracy with which the natives computed time
becomes a subject of prime consideration in a study of their
annals. It would appear that the Eastern Algonkins were
not deficient in astronomical knowledge. Roger Williams
remarks, " they much observe the Starres, and their very
children can give names to many of them ;"^ and the same
testimony is borne by Wassenaer, The latter, speaking of
the tribes around New York Harbor, in 1630, says that their
year began with the first moon after the February moon ; and
that the time for planting was calculated by the rising of the
constellation Taurus in a certain quarter. They named this
constellation the horned head of some great ■ fictitious
animal.*
Zeisberger observes that, in his day, the Lenape did not
have a fixed beginning to their year, but reckoned from one
seeding time to another, or from when the corn was ripe, etc.^
Nevertheless, they had a word for year, gachtin, and counted
their ages and the sequence of events by yearly periods. The
^ E. M. Ruttenber, History of the Indian Tribes of Hudson River,
p. 96, note.
2 Maximilian, Prince of Wied, Travels in America, p. 35.
' A Key into the Language of America, p. 105.
* Documentary Histoiy of Neiv York, Vol. Ill, pp. 29, 32.
5 Grammar of the Language of the Lenni Lenape, pp. 108-109.
56 THE LENAPE AND THEIR LEGENDS.
Chipeways count by winters (^pipun-agak, in which the first
word means winter, and the second is a plural form similar to
the Del. gachfhi); but the Lenape did not apparently follow
them in this. They recognized only twelve moons in the
year and not thirteen, as did the New England nations; at
least, the names of but twelve months have been preserved.^
The day periods were reckoned usually by nights, but it was
not improper to count by "suns" or days.
Pictographic Signs.
The picture writing of the Delawares has been quite fully
described by Zeisberger, Loskiel and Heckewelder. It was
scratched upon stone (Loskiel), or more frequently cut in or
painted upon the bark of trees or pieces of wood. The
colors were chiefly black and red. The system was highly
conventionalized, so that it could readily be understood by
all their tribes, and also by others with whom they came in
contact, the Shawnees, Wyandots, Chipeways, etc.
The subjects had reference not merely to matters of present
interest, but to the former history of their nation, and were
directed "to the preservation of the memory of famous men,
and to the recollection of events and actions of note."
Therefore, their Agamemnons felt no anxiety for the absence
of a Homer, but "confidently reckoned that their noble
deeds would be held in memory long after their bodies had
perished."^
The material on which the drawings were made was
1 They are given, with translations, in Zeisberger's Grammar, p. 109.
2 See Loskiel, Geschichte der Mission, etc., pp. 32, t,"}) > Heckewelder,
History of the Indian A^ations, chap. X.
PICTOGRAPHIC SIGNS. 57
generally so perishable that few examples have been left to
us. One, a stone about seven inches long, found in central
New Jersey, has been described and figured by Dr. Abbott. '
It represents an arrow crossing certain straight lines. Several
"gorgets" (smooth stone tablets pierced with holes for
suspension, and probably used for ceremonial purposes),
stone knives and pebbles, showing inscribed marks and lines,
and rude figures, are engraved in Dr. Abbott's book ; others
similar have been seen in Bucks and Berks counties, Pa.
There was a remarkable series of hieroglyphics, some
eighty in number, on a rock at Safe Harbor, on the Susque-
hanna. They have been photographed and described by
Prof T. C. Porter, of Lancaster, but have yet to be care-
fully analyzed.^ From its location, it was probably the
work of the Susquehannocks, and did not belong to the
general system of Algonkin pictography.
If the rude drawings appended to the early treatises as
signatures of native sachems be taken as a guide, little or no
uniformity prevailed in the personal signs. The same chief-
tain would, on various occasions, employ symbols differing so
widely that they have no visible relation.^
^ Dr. Charles C. Abbott, Friiuitive Industry, pp. 71, 207, 347, 379, 384,
390, 391. Dr. Abbott's suggestion that the bird's head seen on several
specimens might represent the totem of the Turkey gens of the Lenape
cannot be well founded, if Heckewelder is correct in saying that their
totemic mark was only the foot of the fowl. Ind. Nations, p. 253.
2 See Proceedings Atner. Philos. Soc, Vol. X.
s The subject is discussed, and comparative drawings of the native
signatures reproduced, by Prof. D. B. Brunner, in his useful work, The
Indians of Berks County, Pa.,^. 68 (Reading, 1881).
E
58 THE LENAPE AND THEIR LEGENDS.
An interesting incident is recorded by Friend John
Richardson when on a visit to William Penn, at his
manor of Pennsburg, in 1701. Penn asked the Indian
interpreter to give him some idea of what the native notion
of God was. The interpreter, at a loss for words, had
recourse to picture writing, and describing a number of
circles, one inside the other, he pointed to the centre of
the innermost and smallest one, and there, "placed, as he
said, by way of representation, the Great Man."^ The ex-
planation was striking and suggestive, and hints at the
meaning of the not infrequent symbol of the concentric
circles.
An alleged piece of Delaware pictography is copied by
Schoolcraft^ from the London Archceologia, Vol. IV. It pur-
ports to be an inscription found on the Muskingum river in
1780, and the interpretation is said to have been supplied by
the celebrated Delaware chief, Captain White Eyes (Coque-
thagechton). As interpreted, it relates to massacres of the
whites by the Delaware chief, Wingenund, in the border war
of 1763-
There is a tissue of errors here. The pictograph, ''drawn
with charcoal and oil on a tree," must have been quite recent,
and is not likely to have referred to events seventeen years
antecedent. There is no evidence that Wingenund took part
\/ in Pontiac's conspiracy, and he was the consistent friend of
1 John Richardson's Diary, quoted in An Account of the Conduct of
the Society of Friends toward the Indian Tribes, pp. 61, 62 (London,
1844).
''■ History and Statistics of the Indian Tribes, Vol. I, plate 47, B, and
pages 353, 354-
RECORD STICKS. 59
the whites.^ Several of the characters are not like Indian
pictographs. And finally, White Eyes, the alleged inter-
preter in 1780, had died at Tuscarawas, two years before,
Nov. loth, 1778 P
Record Sticks.
The Algonkin nations very generally preserved their myths,
their chronicles, and the memory of events, speeches, etc.,
by means of marked sticks. As early as 1646, the Jesuit
missionaries in Canada made use of these to teach their con-
verts the prayers of the Church and their sermons.''
The name applied to these record or tally sticks was,
among the Crees and Chipeways, mas'sinahigan, which is the
common word now for book, but which originally meant " a
piece of wood marked with fire," from the verb viasindkisan,
I imprint a mark upon it with fire, I burn a mark upon it,*
thus indicating the rude beginning of a system of mnemonic
aids. The Lenape words for book, malackhickan, Camp.,
ma/nalekhican Zeis., were probably from the same root.
In later days, instead of burning the marks upon the sticks,
they were painted, the colors as well as the figures having
certain conventional meanings.^
These sticks are described as about six inches in length,
slender, though varying in shape, and tied up in bundles.*
1 "Amiable and benevolent," says Heckewelder, whose life he aided
in saving on one occasion. Indian Nations, p. 285.
2 E. de Schweinitz, Life of Zeisberger, p. 469.
3 Relation des Jesuites, 1646, p. 33.
4 Baraga, A Dictionary of the Otchipwe Language, s. v.
5 For an example, see de Schweinitz, Life of Zeisberger, p. 342.
6 Documentary History of New York, Vol. IV, p. 437.
60 THE LENAPE AND THEIR LEGENDS.
Such bundles are mentioned by the interpreter Conrad Weiser,
as in use in 1748 when he was on his embassy in the Indian
country.^ The expression, " we tied up in bundles," is trans-
lated by Mr. Heckewelder, olumapsid, and a head chief of
the Lenape, usually called Olomipees, was thus named, ap-
parently as preserver of such records.'^ I shall return on a
later page to the precise meaning of this term.
The word signifying to paint was walatiihi, which does not
appear in western dialects, but is found precisely the same in
the Abnaki, where it is given by Rasles, Sramann,^ which,
transliterated into Delaware (where the / is substituted for the
r), would be w' Ia??i'an. From this word came WaUajnilnk,
the name applied by the natives to a tract in New Castle
county, Delaware, since at that locality they procured supplies
of colored earth, which they employed in painting. It means
" the place of paint." *
Roger Williams, describing the New England Indians,
^ Journal of Conrad Weiser, in Early History of Western Penna., p. 16.
2 Trans. Am. Phil. Soc, Vol. IV, p. 384.
' A Dictionary of the Abnaki Latiguage, s. v. Peinture.
^ See ante p. 53. Mr. Francis Vincent, in his History of the State of
Delaware, p. 36 (Phila., 1870), says of the colored earth of that locality,
that it is " a highly argillaceous loam, interspersed with large and frequent
masses of yellow, ochrey clay, some of which are remarkable for fineness
of texture, not unlike lithomarge, and consists of white, yellow, red and
dark blue clay in detached spots."
The Shawnees applied the same word to Paint Creek, which falls into
the Scioto, close to Chilicothe. They named it Alavionee sepee, of which
Paint Creek is a literal rendering. Rev. David Jones, A Journal of Two
Visits to the West Side of the Ohio in 1772 and 177J, p. 50.
RECORD STICKS. 61
speaks of *' Wunnam, their red painting, which they most
delight in, and is both the Barke of the Pine, as also a red
Earth. "1
The word is derived from Narr. wunne, Del, wulii, Chip.
gwanatsch = beautiful, handsome, good, pretty, etc.
The Indian who had artistically bedaubed his skin with red,
ochreous clay, was esteemed in full dress, and delightful to
look upon. Hence the term w«///, fine, pretty, came to be
applied to the paint itself.
The custom of using such sticks, painted or notched, was
by no means peculiar to the Delawares. They were familiar
to the Iroquois, and the early travelers found them in common
employment among the southern tribes.^
As the art advanced, in place of simple sticks, painted or
notohed, wooden tablets came into use, on which the symbols
were scratched or engraved with a sharp flint or knife. Such
are those still in use among the Chipeway, described by Dr.
James as "rude pictures carved on a flat piece of wood ;"^ by
the native Copway, as " board plates ; " * and more precisely
^ Aey into the Language of America, p. 206.
2 Lawson, in his New Account of Carolina, p. iSo, says that the natives
there bore in mind their traditions by means of a " Parcel of Reeds of
different Lengths, with several distinct Marks, known to none but them-
selves." James Adair writes of the Southern Indians : " They count certain
very remarkable things by notched square sticks, which are distributed
among the head warriors and other chieftains of different towns." History
of the Indians, p. 75.
3 Dr. Edwin James, Narrative of John Tanner, p. 341.
* George Copway, Traditional History of the Ojibzvay Nation, pp. 130,
131-
62 ' THE LENAPE AND THEIR LEGENDS.
by Mr. Schoolcraft, as "a tabular piece of wood, covered on
both sides with a series of devices cut between parallel lines." ^
The Chipeway terms applied to these devices or symbols
are, according to Mr. Schoolcraft, kekeetvin, for those in ordi-
nary and common use, and kekeenowin, for those connected
with the mysteries, the " meda worship" and the "great
medicine." Both words are evidently from a radical signify-
ing a mark or sign, appearing in the words given in Baraga's
''Otchipwe Dictionary," kikinawadjiton, I mark it, I put a
certain mark on it, and kikinoamawa, I teach, instruct him.
Moral and Mental Character.
The character of the Delawares was estimated very
differently, even by those who had the best opportunities of
judging. The missionaries are severe upon them. Brainerd
described them as " unspeakably indolent and slothful. They
have little or no ambition or resolution ; not one in a
thousand of them that has the spirit of a man."^ No more
favorable was the opinion of Zeisberger. He speaks of their
alleged bravery with the utmost contempt, and morally he
puts them down as *' the most ordinary and the vilest of
savages."^
Perhaps these worthy missionaries measured them by the
standard of the Christian ideal, by which, alas, we all fall
wofully short.
Certainly, other competent observers report much more
1 Schoolcraft, Indian Tribes, Vol. I, p. 339.
2 Brainerd, Life and Journal, p. 410. ^
3 E. de Schweinitz, Life and Times of Zeisberger, p. 92.
MORAL AND MENTAL CHARACTER. 63
cheerfully. One of the first explorers of the Delaware,
Captain Thomas Young (1634), describes them as " very
well proportioned,, well featured, gentle, tractable and
docile."^
Of their domestic affections, Mr. Heckewelder writes : " I
do not believe that there are any people on earth who are
more attached to their relatives and offspring than these
Indians are." ^
Their action toward the Society of Friends in Pennsylvania
indicates a sense of honor and a respect for pledges which we
might not expect. They had learned and well understood
that the Friends were non-combatants, and as such they
never forgot to spare them, even in the bloody scenes of
border warfare.
"Amidst all the devastating incursions of the Indians in
North America, it is a remarkable fact that no Friend who
stood faithful to his principles in the disuse of all weapons
of war, the cause of which was generally well understood
by the Indians, ever suffered personal molestation from
them."*
The fact that for more than forty years after the founding
of Penn's colony there was not a single murder committed
on a settler by an Indian, itself speaks volumes for their self-
control and moral character. So far from seeking quarrels
1 Mass. Hist. Soc. Colls., 4th series, Vol. IX, where Captain Young's
journal is printed.
2 Hecketvelder MSS. in Amer Phil. Soc. Lib.
3 An Account of the Conduct of the Society of Friends toward the
Indian Tribes, p. 72 (London, 1S44).
^
64 THE LENAPE AND THEIR LEGENDS.
with the whites they extended them friendly aid and
comfort.^
Even after they had become embittered and corrupted by
the gross knavery of the whites (for example, the notorious
" long walk,") and the debasing influence of alcohol, such an
authority as Gen. Wm. H. Harrison could write these words
about the Delawares : "A long and intimate knowledge of
them, in peace and war, as enemies and friends, has left upon
my mind the most favorable impression of their character for
bravery, generosity and fidelity to their engagements."'^
More than this, and from a higher source, could scarcely be
asked.
That intellectually they were by no means deficient is
acknowledged by Brainerd himself. "The children," he
^ The records of my own family furnish an example of this. My
ancestor, William Brinton, arrived in the fall of 1684, and, with his wife
and children, immediately took possession of a grant in the unbroken
wilderness, about twenty miles from Philadelphia. A severe winter set in ;
their food supply was exhausted, and they would probably have perished
but for the assistance of some neighboring lodges of Lenape, who provided
them with food and shelter. It is, therefore, a debt of gratitude which I
owe to this nation to gather its legends, its language, and its memories, so
that they,
" in books recorded,
May, like hoarded
Household words, no more depart !"
* A Discourse ott the Aborigines of the Valley of the Ohio, p. 25 (Cinn.,
1838). I add the further testimony of John Brickell, who was a captive
among them from 1791 to 1796. He speaks of them as fairly virtuous and
temperate, and adds : " Honesty, bravery and hospitality are cardinal
virtues among them." Narrative of Captivity af/iong the Delaware
Indians, in the American Pioneer, Vol. I, p. 48 (Cincinnati, 1844).
;^VwXyu ^K^'^' '^^'^'^^^^^^^^
RELIGIOUS BELIEFS. 65
writes, " learn with surprising readiness; their master tells
me he never had an English school that learned, in general,
so fast."^
Religious Beliefs.
With the hints given us in various authors, it is not difficult
to reconstruct the primitive religious notions of the Delawares.
They resembled closely those of the other Algonkin nations,
and were founded on those general mythical principles which,
in my "Myths of the New World," I have shown existed
widely throughout America. These are, the worship of Light,
especially in its concrete manifestations of fire and the sun ;
of the Four Winds, as typical of the cardinal points, and as
the rain bringers; and of the Totemic Animal.
As the embodiment of Light, some spoke of the sun as a
deity, ^ while their fifth and greatest festival was held in honor
of Fire, which they personified, and called the Grandfather
of all Indian nations. They assigned to it twelve divine
assistants, who were represented by so many actors in the
ceremony, with evident reference to the twelve moons or
months of the year, the fire being a type of the heavenly
blaze, the sun.^
But both Sun and Fire were only material emblems of the
mystery of Light. This was the " body or fountain of deity,"
which Brainerd said they described to him in terms that he
could not clearly understand ; something " all light ;" a being
^ Life a7td Journal, p. 381.
2 " Others imagined the Sun to be the only deity, and that all things
were made by him." David Brainerd, Lfe and Journal, p. 395.
' Loskiel, Geschichte der Mission, p. 55.
66 THE LENAPE AND THEIR LEGENDS.
"in whom the earth, and all things in it, may be seen;" a
"great man, clothed with the day, yea, with the brightest
day, a day of many years, yea, a day of everlasting continu-
ance." From him proceeded, in him were, to him returned,
all things and the souls of all things.
Such was the extraordinary doctrine which a converted
priest of the native religion informed Brainerd was the teach-
ing of the medicine men.^
The familiar Algonkin myth of the " Great Hare," which
I have elsewhere shown to be distinctively a myth of Light,''
was also well known to the Delawares, and they applied to
this animal, also, the appellation of the " Grandfather of
the Indians." ^ Like the fire, the hare was considered their
ancestor, and in both instances the Light was meant, fire
being its symbol, and the word for hare being identical with
that of brightness and light.
As in Mexico and elsewhere, this light or bright ancestor
was the culture hero of their mythology, their pristine in-
structor in the arts, and figured in some of their legends as a
white man, who, in some remote time, visited them from the
east, and brought them their civilization.*
1 desire to lay especial stress on these proofs of Light wor-
ship among the Delawares, for it has an immediate bearing
on several points in the Walam Olum. There are no com-
^ David Brainerd, Ltji and Jouftial, pp. 395, 399.
2 D. G. Brinton, The Myths of the New World, chap, vi ; American
Hero Myths, chap. ii.
* Loskiel, Geschichte der Jlfission, p. 53.
4 He is thus spoken of in Campanius, Accottnt of New Sweden, Book
III, chap. xi. Compare my Myths of the N'ew World, p. 190.
RELIGIOUS BELIEFS. 67
pounds more frequent in that document than those with the
root signifying "light," "brightness," etc. ; and this is one
of the evidences of its authenticity.
Next in order, or rather, parallel with and a part of the
worship of Light, was that of the Four Cardinal Points,
always identified with the Four Winds, the bringers of rain
and sunshine, the rulers of the weather.
"After the strictest inquiry respecting their notions of the
Deity," says David Brainerd, "I find that in ancient times,
before the coming of the white people, some supposed there
were four invisible powers, who presided over the four corners
of the earth. "^
The Montauk Indians of Long Island, a branch of the
Mohegans, also worshiped these four deities, as we are in-
formed by the Rev. Sampson Occura ; ^ and Captain Argoll
found them again in 1616 among the accolents of the Potomac,
close relatives of the Delawares. Their chief told him : "We
have five gods in all ; our chief god appears often unto us in
the form of a mighty great hare ; the other four have no
visible shape, but are indeed the four winds, which keep the
four corners of the earth." ^
These are the fundamental doctrines, the universal credo, of
not only all the Algonkin faiths, but of all or nearly all primi-
tive American religions.
This is very far from the popular conception of Indian
religion, with its " Good Spirit " and "Bad Spirit." Such
1 Brainerd, Life and Journal, p. 395.
2 His statements are in the Colls, of the Mass. Hist. Soc, Vol. X (ist
Series), p. 108.
3 Wm. Strachey, Historie of Travaile into Virginia, p. 98.
68 THE LENAPE AND THEIR LEGENDS.
ideas were not familiar to the native mind. Heckewelder,
Brainerd and Loskiel all assure us in positive terms that the
notion of a bad spirit, a "Devil," was wholly unknown
to the aborigines, and entirely borrowed from the whites.
Nor was the Divinity of Light looked upon as a benefi-
cent father, or anything of that kind. The Indian did not
appeal to him for assistance, as to his totemic and personal
gods.
These were conceived to be in the form of animals, and
various acts of propitiation to them were performed. Such
acts were not a worship of the animals themselves. Brainerd
explains this very correctly when he says: "They do not
suppose a divine power essential to or inhering in these
creatures, but that some invisible beings, not distinguished
from each other by certain names, but only notionally,
communicate to these animals a great power, and so make
these creatures the immediate authors of good to certain
persons. Hence such a creature becomes sacred to the
person to whom he is supposed to be the immediate author
of good, and through him they must worship the invis-
ible powers, though to others he is no more than another
creature." ^
They rarely attempted to set forth the divinity in image.
The rude representation of a human head, cut in wood, small
enough to be carried on the person, or life size on a post, was
their only idol. This was called tusinkJioalican. They also
drew and perhaps carved emblems of their totemic guardian.
Mr. Beatty describes the head chief's home as a long building
of wood: "Over the door a turtle is drawn, which is the
^ Brainerd, Life and Travels, p. 394.
DOCTRINE OF THE SOUL. 69
ensign of this particular tribe. On each door post was cut
the face of a grave old man."^
Occasionally, rude representations of the human head,
chipped out of stone, are exhumed in those parts of Pennsyl-
vania and New Jersey once inhabited by the Lenape.^ These
are doubtless the wsinkhoalican above mentioned.
Doctrine of the Soul.
.There was a general belief in a soul, spirit or immaterial
part of man. For this the native words were tschipey and
tschitschank (in Brainerd, chichuny). The former is derived
from a root signifying to be separate or apart, while the latter
means " the shadow."^
Their doctrine was that after death the soul went south,
where it would enjoy a happy life for a certain term, and
then could return and be born again into the world. In
moments of spiritual illumination it was deemed possible to
^ Charles Beatty, Journal, p. 44.
2 One, about five inches in height, of a tough, argillaceous stone, is
figured and described by Dr. C. C. Abbott, in the American Naturalist,
October, 1882. It was found in New Jersey.
* From the same root, tschip, are derived the Lenape tschipilek, some-
thing strange or wonderful ; tschepsit, a stranger or foreigner ; and tschapiet,
the invocation of spirits. Among the rules agreed upon by Zeisberger's
converted Indians was this : " We will use no tschapiet, or witchcraft,
when hunting." (De Schweinitz, Life of Zeisberger, p. 379.)
The root tschitsch indicates repetition, and applied to the shadow or
spirit of man means as much as his double or counterpart.
A third word for soul was the verbal form w' tellenape^uoagan, " man —
his substance;" but this looks as if it had been manufactured by the mis-
sionaries.
70 THE LENAPE AND THEIR LEGENDS.
recall past existences, and even to remember the happy epoch
passed in the realm of bliss. ^
The path to this abode of the blessed was by the Milky
Way, wherein the opinion of the Delawares coincided with
that of various other American nations, as the Eskimos, on
the north, and the Guaranis of Paraguay, on the south.
The ordinary euphemism to inform a person that his death
was at hand was : "You are about to visit your ancestors;"^
but most observers agree that they were a timorous people,
with none of that contempt of death sometimes assigned
them.^
The Native Priests.
An important class among the Lenape were those called
by the whites doctors, conjurers, or medicine men, who were
really the native priests. They appear to have been of two
schools, the one devoting themselves mainly to divination,
the other to healing.
According to Brainerd, the title of the former among the
Delawares, as among the New England Indians, \i2.'=> powwow ,
a word meaning " a dreamer;" Chip., bawadjagan, a dream;
nind apawe, I dream ; Cree, pawa-miwin, a dream. They
were the interpreters of the dreams of others, and themselves
claimed the power of dreaming truthfully of the future and
the absent.* In their visions their guardian spirit visited
^ Compare Loskiel, Geschichie, pp. 48, 49 ; Brainerd, Life and Journal,
PP- 314, 396, 399» 400.
2 Schweinitz, Life of Zeisberger, p. 472.
3 Heckewelder, MSS., says that he has often heard the lamentable cry,
niatta wingi angeln, "I do not want to die."
4 «As for the Powaws," says the native Mohegan, the Rev. Sampson
Occum, in his account of the Montauk Indians of Long Island, " they
THE NATIVE PRIESTS. 71
them; they became, in their ov/n words, "all light," and
they "could see through men, and knew the thoughts of
their hearts."^ At such times they were also instructed at
what spot the hunters could successfully seek game.
The other school of the priestly class was called, as we are
informed by Mr. Heckewelder, inedeu.'^ This is the same
term which we find in Chipeway as mide {inedaween, School-
craft), and in Cree as mi few, meaning a conjurer, a member
of the Great Medicine Lodge. ^ I suspect the word is from
m'tieh, heart (Chip, k'tde, thy heart), as this organ was con-
sidered the source and centre of life and the emotions, and
is constantly spoken of in a figurative sense in Indian con-
versation and oratory.
Among the natives around New York Bay there was a body
of conjurers who professed great austerity of life. They had
no fixed homes, pretended to absolute continence, and both
exorcised sickness and officiated at the funeral rites. Their
name, as reported by the Dutch, was kitzinacka, which is
evidently Great Snake {gitschi, achkooJz). The interesting
fact is added, that at certain periodical festivals a sacrifice
say they get their art from dreams." ATass. Hist. Soc. Colls., Vol. X,
p. 109. Dr. Trumbull's suggested affinity of povvaw with Cree tdp-
wayoo, he speaks the truth; Nar., taupowanog, wise speakers, is, I
think, correct ; but the latter are secondary senses. They were wise, and
gave true counsel, who could correctly interpret dreams. Compare the
Iroquois katetsetis, to dream; katetsiens, to practice medicine, Indian
fashion. Cuoq, Lexiqiie de la Langue Iroqtioise.
^ David Brainerd, Life and Journal, pp. 400, 401.
2 Hist. Ind. Nations^ p. 280.
^ Hist, and Statistics of the Indian Tribes, Vol. I, p. 358, seq.
72 THE LENAPE AND THEIR LEGENDS.
was prepared, which it was believed was carried off by a huge
serpent.^
When the missionaries came among the Indians, the shrewd
and able natives who had been accustomed to practice on the
credulity of their fellows recognized that the new faith would
destroy their power, and therefore they attacked it vigorously.
Preachers arose among them, and claimed to have had com-
munications from the Great Spirit about all the matters which
the Christian teachers talked of. These native exhorters
fabricated visions and revelations, and displayed symbolic
drawings on deerskins, showing the journey of the soul after
death, the path to heaven, the twelve emetics and purges
which would clean a man of sin, etc.
Such were the famed prophets Papunhank and Wangomen,
who set up as rivals in opposition to David Zeisberger; and
such those who so constantly frustrated the efforts of the
pious Brainerd. Often do both of these self-sacrificing apos-
tles to the Indians complain of the evil influence which such
false teachers exerted among the Delawares.'^
The existence of this class of impostors is significant for
the appreciation of such a document as the Walam Olum.
They were partially acquainted with the Bible history of
1 Wassenaer's Description of the New Netherlands ( 1 631), in Doc. Hist.
of New York, Vol. IH, pp. 28, 40. Other signs of serpent worship were
common among the Lenape. Loskiel states that their cast-off skins were
treasured as possessing wonderful curative powers [Geschichte, ^. 147),
and Brainerd saw an Indian offering supplications to one {Life and
Journal, p. 395).
2 See Brainerd, Life and Journal, pp. 310, 312, 364, 398, 425, etc.,
and E. de Schweinitz, Life of Zeisberger, pp. 265, 332, etc.
RELIGIOUS CEREMONIES. 73
creation ; some had learned to read and write in the mission
schools ; they were eager to imitate the wisdom of the whites,
while at the same time they were intent on claiming authentic
antiquity and originality for all their sayings.
Religious Ceremonies.
The principal sacred ceremony was the dance and accom-
panying song. This was called kanti kanti, from a verbal
found in most Algonkin dialects with the primary meaning
to sing (Abnaki, skan, je danse et chante en meme temps,
Rasles; CxQe,nikam; Chip., ?«^^w, I sing). From this noisy
rite, which seems to have formed a part of all the native cele-
brations, the settlers coined the word cantico, which has sur-
vived and become incorporated into the English tongue.
Zeisberger describes other festivals, some five in number.
The most interesting is that called Machtoga, which he trans-
lates ''to sweat." This was held in honor of " their Grand-
father, the Fire." The number twelve appears in it frequently
as regulating the actions and numbers of the performers.
This had evident reference to the twelve months of the year,
but his description is too vague to allow a satisfactory analysis
of the rite.
CHAPTER IV.
The Literature and Language of the Lenape.
g I. Literature of the Lenape Tongue. — Campanius ; Penn ; Thomas; Zeisberger;
Heckewelder; Roth; Ettwein ; Grube; Dencke ; Luckenbach ; Henry; Vo-
cabularies ; a native letter.
g 2. General Remarks on the Lenape.
g 3. Dialects of the Lenape.
g 4. Special Structure of the Lenape. — The Root and the Theme ; Prefixes ; SufExes ;
Derivatives; Grammatical Notes.
§ I. Literature of the Lenape Tongue.
The first study of the Delaware language was undertaken
by the Rev. Thomas Campanius (Holm), who was chaplain
to the Swedish settlements, 1 642-1 649. He collected a vo-
cabulary, wrote out a number of dialogues in Delaware and
Swedish, and even completed a translation of the Lutheran
catechism into the tongue. The last mentioned was published
in Stockholm, in 1696, through the efforts of his grandson,
under the title, Lutheri Catechismus, Ofwersatt pa Atnerican-
Virginiske Sprdket, i vol., sm. 8vo, pp. 160. On pages 133-
154 it has a Vocabulariimi Barbaro-Virgineorum, and on pages
155-160, Vocabula Mahakuassica. The first is the Delaware
as then current on the lower river, the second the dialect of
the Susquehannocks or Minquas, who frequently visited the
Swedish settlements.
Although he managed to render all the Catechism into
something which looks like Delaware, Campanius' knowledge
of the tongue was exceedingly superficial. Dr. Trumbull
says of his work : " The translator had not learned even so
74
LITERATURE OF THE LENAPE TONGUE. 75
much of the grammar as to distinguish the plural of a noun
or verb from the singular, and knew nothing of the ''transi-
tions" by which the pronouns of the subject and object are
blended with the verb." ^
At the close of his "History of New Sweden," Campanius
adds further linguistic material, including an imaginary con-
versation in Lenape, and the oration of a sachem. It is of
the same character as that found in the Catechism.
After the English occupation very little attention was given
to the tongue beyond what was indispensable to trading.
William Penn, indeed, professed to have acquired a mastery
of it. He writes: "I have made it my business to under-
stand it, that I might not want an interpreter on any occa-
sion."^ But it is evident, from the specimens he gives,
that all he studied was the trader's jargon, which scorned
etymology, syntax and prosody, and was about as near
pure Lenape as pigeon English is to the periods of Ma-
caul ay.
An ample specimen of this jargon is furnished us by Gabriel
Thomas, in his " Historical and Geographical Account of
the Province and Country of Pensilvania ; and of West-New-
Jersey in America," London, 1698, dedicated to Penn.
Thomas tells us that he lived in the country fifteen years,
and supplies, for the convenience of those who propose visit-
ing the province, some forms of conversation, Indian and
English. I subjoin a short specimen, with a brief commen-
tary : —
1 Transactions of the American Philological Association, 1872, p. 158.
2 Penn, Letter to the Free Society of Traders, 1683, Sec. xii.
76 THE LENAPE AND THEIR LEGENDS.
1. Hiiah takoman ? Friend, from whence com'st?
2. Andogowa 7iee weekin. Yonder.
3. Tony ajidogowa kee weekin ? Where yonder?
4. A7-way?noHse. At Arwaymouse.
5. Xeco kee hatah weekin ? What hast got in thy house ?
6. Nee hatah huska weesyouse og I have very fat venison and good
hziska chetena chase og huska strong skins, with very good tur-
orit chekenip. keys.
7. Chingo kee beta ttee chasa ag When wilt thou bring me skins and
yousa etka chekenip ? venison, with turkeys ?
8. Halapa etka nisha kishquicka. To-morrow, or two days hence.
1. Hitah for n'ischu (Mohegan, nitap), my friend; takoman,
Zeis, takomun, from ta, where, k, 2d pers. sing.
2. Andogozva, similar to undachzve, he comes, Heck. ;
nee, pron. possess, ist person ; weekin = wikwam, or wig-
wam. " I come from my house."
3. Tony, = Zeis, tani, where ? kee, pron. possess. 2d person.
4. Arwaymouse was the name of an Indian village, near
Burlington, N. J.
5. Keco, Zeis, koecu, what? hatah, Zeis, hattin, to have.
6. Huska, Zeis, husca, "very, truly;" wees, Zeis, wisu,
fatty flesh, youse, R. W. jous, deer meat; og. Camp., ock,
Zeis, woak and; chetena, Zeis, tschitani, strong; chase, 7j.
chessak, deerskin ; orit, Zeis, wulit, good ; chekenip, Z. tsche-
kenum, turkey.
7. Chingo, Zeis. tschingatsch,\f\\Qn ; be to, Z. peie?i, to bring;
^/^'a, R. W., ka, and.
8. Halapa, Z, alappa, to-morrow; nisha, two; kishquicka,
Z. gischgu, day, gischguik, by day.
The principal authority on the Delaware language is the
Rev. David Zeisberger, the eminent Moravian missionary,
LITERATURE OF THE LENAPE TONGUE. 77
whose long and devoted labors may be accepted as fixing the
standard of the tongue.
Before him, no one had seriously set to work to master the
structure of the language, and reduce it to a uniform orthog-
raphy. With him, it was almost a lifelong study, as for more
than sixty years it engaged his attention. To his devotion
to the cause in which he was engaged, he added considerable
natural talent for languages, and learned to speak, with almost
equal fluency, English, German, Delaware and the Onondaga
and Mohawk dialects of the Iroquois.
The first work he gave to the press was a *' Delaware
Indian- and English Spelling Book for the Schools of the
Mission of the United Brethren," printed in Philadelphia,
1776. As he did not himself see the proofs, he com-
plained that both in its arrangement and typographical
accuracy it was disappointing. Shortly before his death,
in 1806, the second edition appeared, amended in these
respects. A "Hymn Book," in Delaware, which he finished
in 1802, was printed the following year, and the last work
of his life, a translation into Delaware of Lieberkuhn's
"History of Christ," was published at New York in
1821.
These, however, formed but a small part of the manuscript
materials he had prepared on and in the language. The most
important of these were his Delaware Grammar, and his
Dictionary in four languages, English, German, Onondaga
and Delaware.
The MS. of the Grammar was deposited in the Archives of
the Moravian Society at Bethlehem, Pa. A translation of it
was prepared by Mr. Peter Stephen Duponceau, and published
78 THE LENAPE AND THEIR LEGENDS.
in the " Transactions of the American Philosophical Society,"
in 1827.
The quadrilingual dictionary has never been printed. The
MS. was presented, along with others, in 1850, to the library
of Harvard College, where it now is. The volume is an
oblong octavo of 362 pages, containing about 9000 words
in the English and German columns, but not more than half
that number in the Delaware.
A number of other MSS. of Zeisberger are also in that
library, received from the same source. Among these are a
German-Delaware Glossary, containing 51 pages and about
600 words ; a Delaware-German Phrase Book of about 200
pages ; Sermons in Delaware, etc., mostly incomplete studies,
but of considerable value to the student of the tongue.^
Associated with Zeisberger for many years was the genial
Rev. John Heckewelder, so well known for his pleasant
"History of the Indian Nations of Pennsylvania," his in-
terpretations of the Indian names of the State, and his cor-
respondence with Mr. Duponceau. He certainly had a fluent,
practical knowledge of the Delaware, but it has repeatedly
been shown that he lacked analytical power in it, and that
many of his etymologies as well as some of his grammatical
statements are erroneous.
Another competent Lenapist was the Rev. Johannes Roth.
He was born in Prussia in 1726, and educated a Catholic.
Joining the Moravians in 1748, he emigrated to America in
1756, and in 1759 took charge of the missionary station called
1 On the literary works of Zeisberger, see Rev. E. de Schweinitz, Life
of Zeisberger, chap, xlviii, who gives a full account of all the printed
works, but does not describe the MSS.
LITERATURE OF THE LENAPE TONGUE. 79
Schechschiquanuk, on the west bank of the Susquehanna,
opposite and a little below Shesequin, in Bradford county,
Pennsylvania. There he remained until 1772, when, with
his flock, fifty- three in number, he proceeded to the new
Gnadenhiitten, in Ohio. There a son was born to him, the
first white child in the area of the present State of Ohio. In
1774 he returned to Pennsylvania, and after occupying various
pastorates, he died at York, July 22d, 1791.
Roth has left us a most important work, and one hitherto
entirely unknown to bibliographers. He made an especial
study of the Unami dialect of the Lenape, and composed in
it an extensive religious work, of which only the fifth part
remains. It is now in the possession of the American
Philosophical Society, and bears the title : —
EiN Versuch !
der Geschichte unsers Harm u. Heylandes
Jesu Christi
in dass Delawarische iibersetzt der Unami
von der Marter Woche an
bis zur
Himmelfahrt unsers Herrn
im
Yahr 1770 u. 72 zu Tschechschequaniing
an
der Susquehanna.
Wuntschi mesettschawi tipatta lammowewoagan sekauchsianup.
Wulapensuhalinen, Woehowaolan Nihillalijeng mPatamauwoss.
The next page begins, ''Der fiinfte Theil," and § 86, and
proceeds to § 139. It forms a quarto volume, of title, 9 pages
of contents in German and English, and 268 pages of text
80 THE LENAPE AND THEIR LEGENDS.
in Unami, written in a clear hand, with many corrections
and interlineations.
This is the only work known to me as composed distinctively
in the Unami, and its value is proportionately great as pro-
viding the means of studying this, the acknowledged most
cultivated and admired of the Lenape dialects.
It will be the task of some future Lenape scholar to edit its
text and analyze its grammatical forms. But I believe that
Algonkin students will be glad to see at this time an extract
from its pages.
I select § 96, which is the parable of the marriage feast of
the king's son, as given in Matthew xxii, 1-14.
1. Woak Jesus wtabptonalawoll woak lapi nuwuntschi
And Jesus he-spoke-with-them and again he-began
Enendhackewoagannall nelih* woak wtellawoll.
parables them-to and he-said-to-them.
2. Ne Wusakimawoagan Patamauwoss -J jv^oii%chi l
The his-kingdom God it-is-hke
mejauchsid* Sakima, na Quisall mall'nitauwan Witach-
certain king, his-son he-made-for-him mar-
pungewiwuladtpoagan .
riage.
3. Woak wtellallocalan wtallocacannall, wentschitsch nek
And he-sent-out his-servants the-bidding the
Elendpannik lih* Witachpungewiwuladtpoaganniing wentsch-
those-bidden to marriage those-
imcussowoak ; tschuk necamawa schingipawak.
who-were-bidden, but they they-were-unwilling.
4. Woak lapi wtellallocalan pili wtallocacannall woak
And again he-sent-out other servants and
wtella |P^^^"lJ/}; Mauwiloh nan Elendpannik, {^a^}
he-said-to-them those the-bidden
LITERATURE OF THE LENAPE TONGUE. 81
Nolachtiippoagan 'nkischachtuppui, nihillalachkik Wisuheng-
The-feast I-have-made-the-feast, they-are-killed they-fattened-
pannik auwessissak nemaetschi nhillapannick woak weemi
them beasts the-whole I-killed-them and all
ktakocku 'ngischachtiippui, peeltik lih Witachpungkewi-
I-have-finished come to mar-
wuladtpoaganniing.
riage.
5. Tschuk necamawa mattelemawoawollnenni, woak ewak
But they they-esteemed-it-not and went
ika, mejauchsid enda wtakihacanniing, napilli nihillatschi
away certain thither to-his-plantation-place other
f M'hallamawachtowoaganniing )
I Nundauchsowoaganniing j .
to-merchandise-place.
6. Tschuk allende wtahunnawoawoll neca allocacannall
But some they-seized-them those servants
I quochkikimawoawoll I ^^^^ wunihillawoawoll necamawa.
[ popochpoalimawoawoU j
they-beat-them and they-killed-them they.
7. Elinenni na* Sakima pentanke, nannen lachxu,
When the king heard therefore he-was-angry,
woak wtellallokalan Ndopaluwinuwak, woak wunihillawunga
and he-sent-them warriors and he-slew
jok Nehhillowetschik, woak wulusiimen Wtuten'nejuwaowoll.
these murderers, and he-destroyed their-cities.
I woll I
I panni j
8. Nannen wtella \ ■ y nelih wtallocacannall : No
[ panni j
Then he-said-to-them to his-servants The
Witachpungkewiwuladtpoagan khella nkischachtuppui, tschuk
marriage truly I-have-prepared-it but
1 T-i J -1 f attacu uchtapsiwunewo )
nek Elendpannick i ^ ■ 1 ■ ■ „„ ^ r
^ { wtopielgique juwunewo J .
the those-bidden are-not-to-sit-down-worthy.
9. Nowentschi allmussin ikali mengichiingi Ansijall, woak
Therefore go-ye-away thither to-some-places roads and
82 THE LENAPE AND THEIR LEGENDS,
winawammoh lih Witachpungkewiwuladtpoagan ; na natta
ask-ye-them to marriage those
aween kiluwa mechkaweek (oh).
whom ye find.
10. Woak nek AUocacannak iwak ikali menggichiingi
And the servants they-went thither to-some-places
Aneijall, woak mawehawoawoll peschuwoawak na natta
roads and they-brought-them-together those
aween machkawoachtid, Memannungsitschik woak Wewu-
whom they-found-them the-bad-ones and the-
lilossitschik, woak nel* Ehendachpuingkill weemi taephikka-
good-ones and the at-the-tables all they-
wachtinewo.
seated.
11. Nannen mattemikaeuh na Sakima, nek Elendpannik
Then he-entered-in the king the those-bidden
mauwi pennawoawoU, woak wunewoawoU uchtenda me-
he-saw-them and he-saw-him there cer-
jauchsid Lenno, na matta uchtellachquiwon witachpungkewi
tain man the not wearing a marriage
Schakhokquiwan.
12. Woak wtellawoll neli ;* Elanggomellen, ktelgiquiki
And he-said-to-him to-him Friend like ,
matte attemiken jun [or ta elinaquo wentschi jun k'mattl-
not ashamed here not like therefore here thou-art-
mikeen,) ; woak \ .|.i ^ [ mattacu witachpungkewi
ashamed and not marriage
Schakhokquiwan ktellachquiwon ? Necama tschuk k'pettunevi.
coat thou wearest He but he-mouth-shuts.
13. Nannen w'tellawoll na Sakima nelih* Wtallocacan-
Then he-said-to-them the king to-them his-ser-
niing; Kachpiluh j ^^ \ Wunachkall woak W'sittall, woak
vants Fasten-ye-him his-hands and his-feet and
LITERATURE OF THE LENAPE TONGUE. 83
lannehewik quatschemung enda achwipegniink, nitschlenda
throw-him where in pitch-darkness even-some
Lipackcuwoagan woak Tschsetschak koalochinen.
weeping and teeth-gnashing
14. Ntitechquoh macheli moetschi wentschimcussuwak,
Because many they-are-called
tschuk tatthiluwak achriceknuksitschik.
but they-are-few the-chosen.
The asterisk occurs in the original apparently to indicate
that a word is superfluous or doubtful. The interlined
translation I have supplied from the materials in the mission-
Delaware dialect, but my resources have not been sufficient
to analyze each word ; and this, indeed, is not necessary for
my purpose, which is merely to present an example of the
true Unami dialect.
The Moravian Bishop, John Ettwein, was another of their
fraternity who applied himself to the study of the Delaware.
Born in Europe in 1712, he came to the New World in 1754,
and died at the great age of ninety years in 1802. He pre-
pared a small dictionary and phrase book, especially rich in
verbal forms. It is an octavo MS. of 88 pages, without title,
and comprises about 1300 entries. This manuscript exists in
one copy only, in the Moravian Archives at Bethlehem.
Bishop Ettwein also prepared for General Washington, in
1788, an account of the traditions and language of the natives,
including a vocabulary. This was found among the Wash-
ington papers by Mr. Jared Sparks, and was published in the
"Bulletin of the Pennsylvania Historical Society," 1848.
One of the most laborious of the Moravian missionaries
was the Rev. Adam Grube. His life spanned nearly a cen-
tury, from 1 715, when he was born in Germany, until 1808,
84 THE LENAPE AND THEIR LEGENDS.
when he died in Bethlehem, Pa. Many years of this were
spent among the Delawares in Pennsylvania and Ohio. He
was familiar with their language, but the only evidence of his
study of it that has come to my knowledge is a MS. in the
Harvard College Library, entitled, "Einige Delawarische
Redensarten und Worte." It has seventy-five useful leaves,
the entries without alphabetic arrangement, some of the verbs
accompanied by partial inflections. The only date it bears
is "Oct. lo, 1800," when he presented it to the Rev. Mr.
Luckenbach, soon to be mentioned.
After the War of 181 2 the Moravian brother, Rev. C. F.
Dencke, who, ten years before had attempted to teach the
Gospel to the Chipeways, gathered together the scattered
converts among the Delawares at New Fairfield, Canada West.
In 1 818 he completed and forwarded to the Publication Board
of the American Bible Society a translation of the Epistles
of John, which was published the same year.
He also stated to the Board that at that time he had finished
a translation of John's Gospel and commenced that of Matthew,
both of which he expected to send to the Board in that year.
A donation of one hundred dollars was made to him to en-
courage him in his work, but for some reason the prosecution
of his labors was suspended, and the translation of the Gospels
never appeared (contrary to the statements in some biblio-
graphies).
It is probable that Mr. Dencke was the compiler of the
Delaware Dictionary which is preserved in the Moravian
Archives at Bethlehem. The MS. is an oblong octavo, in a
fine, but beautifully clear hand, and comprises about 3700
words. The handwriting is that of the late Rev. Mr. Kamp-
LITERATURE OF THE LENAPE TONGUE. 85
man, from 1840 to 1842 missionary to the Delawares on the
Canada Reservation. On inquiring the circumstances con-
nected with this MS., he stated to me that it was written at
the period named, and was a copy of some older work, pro-
bably by Mr. Dencke, but of this he was not certain.
While the greater part of this dictionary is identical in
words and rendering with the second edition of Zeisberger's
"Spelling Book" (with which I have carefully compared it),
it also includes a number of other words, and the whole is
arranged in accurate alphabetical order.
Mr. Dencke also prepared a grammar of the Delaware, as
I am informed by his old personal friend. Rev. F. R. Holland,
of Hope, Indiana ; but the most persistent inquiry through
residents at Salem, N. C, where he died in 1839, and at the
Missionary Archives at Bethlehem, Pa., and Moraviantown,
Canada, have failed to furnish me a clue to its whereabouts.
I fear that this precious document was '' sold as paper stock,"
as I am informed were most of the MSS. which he left at his
decease ! A sad instance of the total absence of intelligent
interest in such subjects in our country.
The Rev. Abraham Luckenbach may be called the last of
the Moravian Lenapists. With him, in 1854, died out the
traditions of native philology. Born in 1777, in Lehigh
county, Pennsylvania, he became a missionary among the
Indians in 1800, and until his retirement, forty-three years
later, was a zealous pastor to his flock on the White river,
Indiana, and later, on the Canada Reservation. His pub-
lished work is entitled " Forty-six Select Scripture Narratives
from the Old Testament, embellished with Engravings, for
the Use of Indian Youth. Translated into Delaware Indian,
86 THE LENAPE AND THEIR LEGENDS.
by A. Luckenbach. New York. Printed by Daniel Fan-
shaw, 1838," Svo, pp. xvi, 304.
After his retirement in Bethlehem, he edited, in 1847, the
second edition of Zeisberger's " Collection of Hymns," the
first of which has already been mentioned.
A short MS. vocabulary, in German and Delaware, is in
the possession of his family, in Bethlehem, and some loose
papers in the language.
One of the most recent students of the Delaware was Mr.
Matthew G. Henry, of Philadelphia. In 1859 and i860 he
compiled, with no little labor, a " Delaware Indian Diction-
ary," the MS. of which, in the library of the American
Philosophical Society, forms a thick quarto volume of 843
pages, with a number of maps. It is in three parts: i,
English and Delaware ; 2, Delaware and English; 3, Dela-
ware Proper Names and their Translations.
It includes, without analysis or correction, the words in
Zeisberger's "Spelling Book," Roger William's "Key,"
Campanius' Vocabulary, those in Smith's and Strachey's
"Virginia," and various Nanticoke, Mohegan, Minsi and
other vocabularies. The derivations of the proper names are
chiefly from Heckewelder, and in other cases are venture-
some. The compilation, therefore, while often useful, lacks
the salutary check of a critical, grammatical erudition, and
in its present form is of limited value.
Some of the later vocabularies collected by various trav-
elers offer points for comparison, and may be mentioned here.
In 1786 Major Denny,^ at Fort Mcintosh, Ohio, collected
1 Major Ebenezer Denny's "Journal," in Memoirs of the Hist. Soc. of
Penna., Vol. VH, pp. 481-86.
LITERATURE OF THE LENAPE TONGUE. 87
a number of Delaware words, principally from Shawnee
Indians. A comparison shows many of them to be in a
corrupt form, owing either to the ignorance of the Shawnee
authority, or to the inaccuracy of Major Denny in catching
the sounds.
While engaged on the Pacific Railroad survey, in 1853,
Lieut. Whipple^ collected a vocabulary of a little over 200
words from a Delaware chief, named Black Beaver, in the
Indian Territory, which was edited, in 1856, by Prof. Turner.
It is evidently a pure specimen, and, as the editor observes,
"agrees remarkably" with earlier authentic vocabularies.
In the second volume of Schoolcraft's large work^ is a
vocabulary of about 350 words, obtained by Mr. Cummings,
U. S. Indian Agent. The precise source, date and locality
are not given, but it is evidently from some trustworthy
native, and is quite correct.
Some small works for the schools of the Baptist missions
among the Delawares in Kansas were prepared by the Rev.
J. Meeker. They appear to be entirely elementary in char-
acter.
It will be observed that in this list not a single native writer
is named. So far as I have ascertained, though many learned
to write their native tongue, not one attempted any composi-
tion in it beyond the needs of daily life.
To make some amends for this, and as I wished to obtain
an example of the Lenape of to-day, I asked Chief Gottlieb
Tobias, an educated native on the Moravian Reservation in
^ Report upon the Indian Tribes, by Whipple, Evvbank and Turner,
p. 56 (Washington, 1855).
* History and Statistics of the Indian Tribes, Vol. II, p. 470.
88 THE LENAPE AND THEIR LEGENDS.
Canada, to give me in writing his opinion of the Delaware
text of the Walum Olum, which I had sent him. This he
obligingly did, and added a translation of his letter. The
two are as follows, without alteration : —
MORAVIANTOWN, Sept. 26, 1 884.
I, Gottlieb Tobias,
Nanne ni ngutschi nachguttemin, jun awen eat ma elekhigetup.
Woak alende nenostamen woak alende taku all wtallichsin ale-
wondasik wiwonalatokowo pachsi wonamii lichsu woak pachsi pilli
lichsoagan. Taku ni nanostamowin. Lamoa namochomsinga
achpami aat nawinachka woak chash tichi kachtin nbibindamaneb
nin lichsoagan. Mauchso lenno woak mauchso chauchshissis woak
juque mauchso chauchshissis achpo pomauchsu igabtshi lua wi-
wonallatokowo won bambil alachshe. Woak lua lamoa ni anda.
Mimansiane ntelsitam alowi ayachichson won alhagawit woak
ehelop ne likhiqui. Gichgi wonami lichso shuk tatcamse woak
gichgi minsiwi lichso.
TRANSLATION.
Than I will try to answer this (which) soma one at some time
wrote. And soma I understand, and soma not, because his language
is called Wonalatoko, half Unami and half another language. I
do not understand it. Long ago my grandfather about 48 years
ago I heard it that language. One man and one old woman and
now another old woman here lives yet who uses this Wonalatoko
language just like this book and she said, I of old time when I was
a child heard more difficult dialect than the present, and many at
that time partly Unami he speak, but sometimes also partly
Minsi he speak.
The drift of Chief Tobias' letter is highly important to
this present work, though his expressions are not couched in
GENERAL REMARKS ON THE LENAPE. 89
the most perfect English. It will be noted that he recognizes
the text of the Walum Olum to be a native production com-
posed in one of the ancient southern dialects of the tongue,
the Unami (Wonami) or the Unalachtgo (Wonalatoko). I
shall recur to this when discussing the authenticity of that
document on a later page.
§ 2. General Remarks on the Lenape.
The Lenape language is a well-defined and quite pure
member of the great Algonkin stock, revealing markedly the
linguistic traits of this group, and standing philologically, as
well as geographically, between the Micmac of the extreme
east and the Chipeway of the far West.
These linguistic traits, common to the whole stock, I may
briefly enumerate as follows : —
1. All words are derived from simple, monosyllabic roots,
by means of affixes and suffixes.
2. The words do not come within the grammatical cate-
gories of the Aryan language, as nouns, adjectives, verbs and
other ''parts of speech," but are "indifferent themes," which
may be used at will as one or the other. To this there
appear to be a few exceptions.
3. Expressions of being (z. e., nominal themes) undergo
modifications depending on the ontological conception as
to whether the thing spoken of is a living or a lifeless
object. This forms the "animate and inanimate," or the
"noble and ignoble" declensions and conjugations. The
distinction is not strictly logical, but largely grammatical,
many lifeless objects being considered living, and the
reverse. This is the only modification of the kind known,
90 THE LENAPE AND THEIR LEGENDS.
true grammatical gender not appearing in any of these
tongues.
4. Expressions of action (/. e., verbal themes) undergo
modifications depending on the abstract assumption as to
whether the action is real or conjectural. If the latter, it is
indicated by a change in the vowel of the root. This leads
to a fundamental division of verbal modes into positive and
suppositive modes.
5. The expression of action is subordinate to that of being,
so that the verbal elements of a proposition are secondary to
the nominal or pronominal elements, and the subjective rela-
tion becomes closely akin to, or identical with, that of pos-
session.^
6. The conception of number is feebly developed in its
application to inanimate objects, which often have no gram-
matical plurals. The inclusive and exclusive plurals are used
in the first person.
7. The genius of the language is holophrastic — that is, its
effort is to express the relationship of several ideas by com-
bining them in one word. This is displayed : i, in nominal
themes, by poly synthesis, by which several such themes are
1 I am aware that in this proposition I am following the German and
French linguists, Steinthal, F. Miiller, Adam, Henry, etc., and not our
own distinguished authority on Algonkin grammar, Dr. J. Hammond
Trumbull, who, in his essay " On the Algonkin Verb," has learnedly
maintained another opinion ( Transactions of the American Philological
Associatiott, 1876, p. 146). I have not been able, however, to convince
myself that his position is correct. The formative elements of the Algon-
kin paradigms appear to me simply attached particles, and not true
inflections. Their real character is obscured by phonetic laws, just as
in the Finnish when compared with the Hungarian.
DIALECTS OF THE LENAPE. 91
welded into one, according to fixed laws of elision and
euphony; and 2, by incorporation, where the object (or a
pronoun representing it) and the subject are united with the
verb, forming the so-called "transitions," or ''objective
conjugations."
8. There is no relative pronoun, so that the relation of
minor to major clauses is left to be indicated either by posi-
tion or the offices of a simple connective.
9. The language of both sexes is identical, those differ-
ences of speech between the males and females, so frequently
observed in other American tongues, finding no place in the
Algonkin.
10. No independent verb-substantive is found, and, as
might be anticipated, no means of predicating existence
apart from quality and attribute.
§ 3. Dialects of the Lenape.
Two slightly different dialects prevailed among the Dela-
wares themselves, the one spoken by the Unami and Una-
lachtgo, the other by the Minsi. The former is stated by the
Moravian missionaries to have had an uncommonly soft and
pleasant sound to the ear,^ and William Penn made the same
remark. It was also considered to be the purer and more
elegant dialect, and was preferred by the missionaries as the
vehicle for their translations.
The Minsi was harsher and more difficult to learn, but
1 "Ungemein wohlklingend." Loskiel, Geschichte der Mission, p. 24.
An early traveler of English nationality pronounced it " sweet, of noble
sound and accent." Gabriel Thomas, Hist, and Geog. Account of Pen-
silvania and West New Jersey, p. 47 (London, 1698).
92
THE LENAPE AND THEIR LEGENDS.
would seem to have been the more archaic branch, as it is
stated to be a key to the other, and to preserve many words
in their integrity and original form, which in the Unami were
abbreviated or altogether dropped. The Minsi dialect was
closely akin to the Mohegan.
How far the separation of the Delaware dialects had ex-
tended may be judged from the subjoined list of words. They
are selected, as showing the greatest variation, from a list of
over one hundred, prepared by Mr. Heckewelder for the
American Philosophical Society, and preserved in MS. in its
library.
The comparison proves that the differences are far from
extensive, and chiefly result from a greater use of gutturals.
COMPARISON
OF THE UNAMI AND
MINSI DIALECTS.
Unami.
Minsi.
God
Patamawos
Pachtamawos
Earth
hacki
achgi
Valley
pasaeck
pachsajech
Beard
wuttoney
wuchtoney
Tooth
wipit
wichpit
Blood
mocum
mochcum
Night
ipocu
ipochcu
Pretty
schiki
pschickki
Small
tangeto
tschankschisu
Stone
assinn
achsiin
The Sea
kithanne
gichthanne
Light
woacheu
woashe^jeek
Black
siiksit
neesachgissit
Chief
saki^ma
wajauwe
Green
asgask
asgasku
No, not
matta
machta
DIALECTS OF THE LENAPE.
93
What differences there were have been retained and perhaps
accentuated in modern times, if we may judge from the names
of consanguinity obtained by Mr. Lewis H, Morgan on the
Kansas Reservation in i860. These are given in part in the
annexed table, and the Mohegan is added for the sake of ex-
tending the comparison.
Delaware.
Minsi.
Mohegan.
My grandfather
nu mohomus
na niahomis''
nuh mahome''
My grandmother
noo home''
na nohome
no ome''
My father
noh^h
na no^uh
noh
My mother
nga'hase
nain guk''
n'guk
My son
n'kweese''
nain,gwase''
n'diome''
My daughter
n^danuss
nain daness''
ne chune''
My grandchild
noh whese''
nain no whas6
na hise''
My elder brother
nah hans
nain n^hans
n ta kun''
My elder sister
na mese''
nain nawase
na mees
My younger brother
nah eese umiss
nain hisesamus''
nhisum
A noteworthy difference in the Northern and Southern
Lenape dialects was that the latter possessed the three pho-
netic elements n, I and r, while the former could not pro-
nounce the r, and their neighbors, the Mohegans, neither the
/ nor the r.
The dialect studied by Campanius and Penn, and that
in southern New Jersey presented the r sound where the
Upper Unami and Minsi had the /. Thus Campanius gives
rheniis, for lenno, man ; and Penn oret, for the Unami wulit,
good.
The dialectic substitution of one of these elements for
another is a widespread characteristic of Algonkin phonology.
94 THE LENAPE AND THEIR LEGENDS.
Roger Williams early called attention to it among the tribes
of New England.^
Tracing it to its origin, it clearly arises from the use of
"alternating consonants," so extensive in American lan-
guages. In very many of them it is optional with the speaker
to employ any one of several sounds of the same class. This
is the case with these letters in Cree, which, for various
reasons, may be considered the most archaic of all the Algon-
kin dialects. In its phonetics, the th, y, I, n and r are ''per-
muting" or "alternating" letters.^
Often, too, the sound falls between these letters, so that
the foreign ear is left in doubt which to write.
That this is the case with the Delaware is evident from
some of the more recent vocabularies where the r is not
infrequent. The following words, from the vocabulary in
Major Denny's Memoir, illustrate this :. —
Stone seegriana
Buffalo serelea
Beaver ihomagru
Above , hoqru7iog, etc.
Even Mr. Lewis A. Morgan, who had considerable practice
in writing the sounds of the Indian languages, inserts the r
in a number of pure Delaware words he collected in Kansas.^
Another difficulty presents itself in the sibilants. They
are not always distinguished.
1 Key into the Language of North America, p. 129. See, also, Mr.
Pickering's remarks on the same subject, in his Appendix to Rasles'
Dictionary of the Abnaki.
2 Howse, Gramniar of the Cree Language, p. 316.
5 See his Ancient Society, pp. 172-73.
DIALECTS OF THE LENAPE. 95
Mr. Horatio Hale writes me on this point : " In Minsi,
and perhaps in all the Lenape dialects, the sound written s is
intermediate between s and M (the Greek (9). This element
is pronounced by placing the tongue and teeth in the position
of the theta, and then endeavoring to utter j."
The guttural, represented in the Moravian vocabularies by
c/i, was softened by the English likewise to the s sound, as
it appears also to have been by the New Jersey tribes.^
In connection with dialectic variation, the interesting ques-
tion arises as to the rapidity of change in language. With
regard to the Lenape we are enabled to compare this for a
period covering more than two centuries. To test it, I have
arranged the subjoined table of words culled from three writers
at about equidistant points in this period. Each wrote in the
orthography of his own tongue, and this I have not altered.
The words from Campanius are from the southern dialect,
which preferred the r to the /, and this substitution should be
allowed for in a fair comparison.
^ The native name of William Penn offers an instance of this phonetic
alteration. It is given as Onas. The proper form is Wonach. It literally
means the tip or extremity of anything ; as ivonach-sitall, the tips of the
toes; wonach-gnlinschall, the tips of the fingers. The inanimate plural
form wolanniall, means the tail feathers of a bird. To explain the name
Penn to the Indians a feather was shown them, probably a quill pen, and
hence they gave the translation Wonach, corrupted into Onas.
96
THE LENAPE AND THEIR LEGENDS.
COMPARISON OF THE DELAWARE AT INTERVALS DURING 2IO
YEARS.
Campmiius.
Zeisberger.
Whipple.
1645.
1778.
1855.
<
Swedish Orthography.
German Orthography.
English Orthography,
Man
rhenus
lenno
lenno
Woman
aquaeo
ochque
h'que'i
Father
nwk
nooch (my)
nuuh
Mother
kahaess
gahowes
gaiez
Head
kwijl
wil
wil
Hair
mijrack
milach
milakh
Ear
hittaock
w'hittawak (pi.)
howitow
Eye
schinck
w'ushgink
tukque^ling
Nose
wikiiwan
w'ikiwan
ouiki'o
Mouth
twn
w'doon
ouitun
Tongue
hijrano
w'ilano
ouilano
Tooth
wippit
w'epit
ouipita
Hand
alaenskan
w'anach
puck-alenge
Foot
zijt
sit
zit
Heart
chitto, kitte
ktee (thy)
hute
House
wickwmen
wiquoam
ouigwam
Pipe
hopockan
hopenican
haboca
Sun
chisogh
gischuch
kishu'h
Star
aranck
alank
alanq'
Fire
taenda
tindey
tundaih
Water
bij
mbi
bih
Snow
kuun
guhn
ku^no
COMPARISON OF
DELAWARE NUMERALS.
Campanms. Thomas.
Zeisberger.
Whipple.
1645.
1695.
I7SO.
1855.
I Ciutte
Kooty
Ngutti
Co^te
2 Nissa
Nisha
Nischa
Ni'sha
3 Naha
Natcha
Nacha
Naha^
4 Nasvvo
Neo
Newo
Ne'ewah
5 Pareenach Pelenach
Palenach
Pahle'nah'k
6 Ciuttas
Kootash
Guttasch
Cot^tasch
7 Nissas
Nishash
Nischasch
Ni^shasch
8 Haas
Choesh
Chasch
Hasch
9 Paeschum Peshonk
Peschkonk
Pes'co
10 Thccrer
L Telen
Tellen
Te^en
DIALECTS OF THE LENAPE. 97
I have no doubt that if a Swede, a German and an English-
man were to-day to take down these words from the mouth
of a Delaware Indian, each writing them in the orthography
of his own tongue, the variations would be as numerous as
in the above list, except, perhaps, the ancient and now dis-
used r sound. The comparison goes to show that there has
probably been but a very slight change in the Delaware, in
spite of the many migrations and disturbances they have
undergone. They speak the language of their forefathers as
closely as do the English, although no written documents
have aided them in keeping it alive. This is but another
proof added to an already long list, showing that the belief
that American languages undergo rapid changes is an error.
The dialect which the Moravian missionaries learned, and
in which they composed their works, was that of the Lehigh
Valley. That it was not an impure Minsi mixed with Mo-
hegan, as Dr. Trumbull seems to think,^ is evident from the
direct statements of the missionaries themselves, as well as
from Heckewelder's Minsi vocabularies, which show many
points of divergence from the printed books. Moreover,
among the first converts from the Delaware nation were
members of the Unami or Turtle tribe, and Zeisberger was
brought into immediate contact with them.^ We may fairly
consider it to have been the upper or inland Unami, which,
as I have said, was recognized by the nation as the purest,
or at least the most polished dialect of their tongue. It
stood midway between the Unalachtgo and Southern Unami
and the true Minsi.
1 Trans. Ain. Philol. Assoc, 1872, p. 157.
^ De Schweinitz, Life of Zeisberger, p. 131.
98 THE LENAPE AND THEIR LEGENDS.
§ 4. Special Structure of the Lenape.
The Root and the Formation of the Theme. — As they appear
in the language of to-day, the Lenape radicals are chiefly
monosyllables, which undergo more or less modifications
in composition. They cannot be used alone, the tongue
having long since passed from that interjectional condition
where each of these roots conveyed a whole sentence in
itself.
Whether they can be resolved back into a few elementary
sounds, primitive elements of speech, I shall not discuss.
This has been done for the Cree roots by Mr. Joseph Howse,^
and most of the radicals of that tongue are identical with
those of the Lenape. Some of his conclusions appear to me
hazardous and hypothetical ; and certainly many of his sup-
posed analogies drawn from European tongues are extrava-
gant.
As in other idioms, so in Lenape, two or more radicals
may be compounded to form a combination, which, in
turn, performs the offices of a radical in the construction of
themes.
This combination is formed either by prefixes or suffixes.
The prefixes are generally adjectival in signification, while
the suffixes are usually classificatory. A number of these are
secondary roots, which are themselves capable of further
analysis.
As so much of the strength of the languages depends on
this plan of word building, I have drawn off" a list of a few
1 A Grammar of the Cree Language , with which is combined an
Analysis of the Chippeway Dialect, by Joseph Hovvse, Esq. (London,
1844)-
LENAPE SUFFIXES. 99
of the more frequent affixes of the Lenape, with their signifi-
cation: —
Lenape Prefixes.
awoss-, beyond, the other side of.
eluwi-, most, a superlative form.
gisch-, see page 102,
kit-, great, large.
lappi-, again, indicates repetition.
lenno-, male, man,
lippoe-, wise, shrewd ; as Uppoeweno, a shrewd man.
mack-, evil, bad, hurt.
matt-, negative and depreciatory ; as mattaptonen, to speak un-
civilly.
ni-, see page loi.
ochque-, she, female.
pack-, division, separation ; pachican, a knife ; pacJiat, to split.
pal-, negative, as dis- or in-, from palli, otherwheres.
tach-, pairs or doubles.
tschitsch-, indicates repetition.
wit-, with or in common.
wul-, or wel-, see page 104.
Many of these are abbreviated to the extent that a single
significant letter is all that remains, as min in msim, hickory
nut ; pakihm, cranberry ; and so acki to k, hanne to an, as
kitanink (Kittanning), from gitscM, great; hanne, flowing
river; ink, locative, "at the place of the great river."
Lenape Suffixes,
-ak, wood, from tachan ; kuwenchak, pine wood.
-aki, place, land.
-ammen, acceptance, adoption ; 'wulista7ne7i, I accept it as
good, I believe it. See page 104.
100 THE LENAPE AND THEIR LEGENDS.
-ape, male, man. From a root ap, to cover (carnally). In
Chipeway applied only to lower animals.
-atton, or hatton, to have, to put somewhere. The radical is at.
Also a prefix, as, hattape, the bow ; lit., what the man has.
-bi, tree ; inachtschibi, papaw tree.
-chtim, a quadruped.
-elendam, a verbal termination, signifying a disposition of mind.
The root is en, ne, ni, I ; "it is to me so."
-gook, a snake ; from achgook, a serpent.
-hanna, properly hannek, a river ; from the root, which appears
in Cree as anask, to stretch out along the ground ; mech-
hannek, a large stream.
Heckewelder derives this from amhamme, a river. The terminal k is, how-
ever, part of the root, and not the locative termination. The word is allied to
Del. guenek, long.
-kikan, tidal water ; kittahikan, the ocean ; shajahikan, the sea
shore.
-hilleu, it is so, it is true ; impersonal form from lissijt.
-hittuck, river, water in motion.
-igan, instrumental ; also sh'ican and can. A participial termi-
nation used with inanimate objects.
-in or ini, of the kind; like; predicative form of the demon-
strative pronoun.
-ink or unk, place where.
-is or -it, diminutive termination.
-leu, it is so, it is true.
-meek, a fish ; viaschila7nek, a trout.
-min, a fruit.
-peek, a body of still water ; inenuppek, a lake.
-sacunk, an outlet of a stream into another ; also saqidk.
-sipu, stream ; lit., stretched, extended.
-tift, with, or in common.
-///, diminutive termination ; atnentit, a babe.
EXAMPLES OF LENAPE DERIVATIVES. 101
-wagan, abstract verbal termination ; machelemuxoivagan, the
being honored.
-wehelleu, a bird.
-wi, the verb-substantive termination, predicating being ; tehek,
cold ; tchekwi, he or it is cold.
-wi, negative termination in certain verbal forms.
-xit, indicates the passive recipient of the action ; machelemuxit,
the one who is honored.
The analysis of a series of derivatives from the same root
offers a most instructive subject for investigation in the Lenape.
Not only does it reveal the linguistic processes adapted, but
it discloses the psychology of the native mind, and teaches
us the associations of its ideas, and the range of its imagina-
tive powers. By no other avenue can we gain access to the
intimate thought-life of this people. Here it is unfolded to
us by evidence which is irrefragable.
These considerations lead me to present a few examples of
the derivatives from roots of different classes.
EXAMPLES OF LENAPE DERIVATIVES.
Subjective Root Ni, /, mitie.
1. In a good sense.
Nihilleu, it is I, or, mine.
Nihillatschi, self, oneself.
Nihillapewi, free {ape, man = I am my own man).
Nihillapewit, a freeman.
Nihillasoivagan, freedom, liberty.
Nihillapeuhen, to make free, to redeem.
Nihillapeuhoalid, the Redeemer, the Saviour.
2. In a bad sense.
Ni'hillan, he is mine to beat, I beat him.
NihiVlan, I beat him to death, I kill him.
102 THE LENAPE AND THEIR LEGENDS.
Nihillowen, I put him to death, I murder him.
Nihillowet, a murderer.
Nihilloivewi, murderous.
3. In a demonstrative sense.
Ne, pi. nek, or ?ien, this, that, the.
Nail, nan, nanne, nanni, this one, that one.
Nill, these.
Naninga, those gone, with reference to the dead.
4. In a possessive sense.
Nitaton, in-my-having, I can, I am able, I know how.
Nitaus, of-my-family, sister-in-law.
Nitis, of-mine, a friend, a companion.
Nitsch ! my child ! exclamation of fondness.
The strangely conflicting ideas evolved from this root
already attracted the attention of Mr. Duponceau.^ That
the notions for freedom and servitude, murderer and Saviour,
should be expressed by modifications of the same radical is
indeed striking ! But the psychological process through which
it came about is evident on studying the above arrangement.
Objective-intejisive root gisch or kich {Cree, Kis or kik).
Signification — successful action.
I. Applied to persons.
A. Initial successful action.
Gischigin, to begin life, to be born.
Gischihan, to form, to make with the hands.
Gischiton, to make ready, to prepare.
Gischeleman, to create with the mind, to fancy,
Gischelendafn, to meditate a plan, to lie.
B. Continuous successful action.
Gischikenamen, to increase, to produce fruit.
^ In a note to Zeisb'erger's Grammar of the Delaware, p. 141.
EXAMPLES OF LENAPE DERIVATIVES. 103
Giken, to grow better in health.
Gikeowagan, life, health.
Gikey, long-living, old, aged.
C. Final successful action.
Gischatten, finished, ready, done, cooked.
Gischito7t, to make ready, to finish.
Gischpuen, to have eaten enough.
Gischileii, it has proved true.
Gischatschimohin, to have resolved, to have decreed.
Gischachpoanhe , baked, cooked (the bread is).
2. Applied to things.
A. Initial successful action.
Gischiech, sun, moon, day, month. The idea appears
to be the beginning of a period of time, with the col-
lateral notion of prosperous activity. The correctness
of the derivation is shown by the next word.
Gischapan, day-break, beginning day-light. From
wapan, the east, or light.
Gischtichwipall, the rays of the sun.
Gischcu, or Gischqiiik, day.
B. Continuous successful action.
Gischten, clear, light, shining.
Gischachsummen, to shine, to enlighten.
Gisc/mten, warm, tepid.
Numerous other derivatives could be added, but the above
are sufficient to show the direction of thoughts flowing from
this root. Howse considers it identical with the root kikh,
great, large. ^ This would greatly increase its derivatives.
They certainly appear allied. In Cree, Lacombe gives kiUhi,
^ A Grammar of the Cree Language, p. 175.
104 THE LENAPE AND THEIR LEGENDS.
great, and kije, finished, perfect, both being terms applied to
divinity.^
General Algonkin root 8 j N 1 1. Abnaki, 8ri ; Micmac, Be^li,
Chippeway, gwan-; Del, two for7ns, WUL attd win. //
conveys the idea of pleasurable sensation.
A. First form, will.
Wulit, well, good, handsome, fine.
Wullihilleu, it is good, etc.
Wuliken, it grows well.
Wulamoe, he truth- speaks.
Wulamoewagan, truth.
Widistatnen, to believe, to accept as truth.
Wnlenensm, to be fine in appearance, to dress.
Wiilene7isen, to be fine to oneself, to be proud.
B. Second form, won or win.
Wimt, ripe, good to eat.
Wonita, he is ripe for it, he can, he is able.
Wingan, sweet, savory.
Winktek, done, boiled, fit to eat.
Winak, sassafras. From its sweet leaves.
Wingz, gladly, willingly.
Winginamen, to delight in.
The figure 8 in the above represents the "whistled a:/,"
like the wh in "which," when strongly pronounced.
From this root, as I have already said, is derived also the
word Walam, red paint, from the sense "to be fine in appear-
ance, to dress," as the Indian accomplished that object by
painting himself.
^ Dictionnaire de la Langue des Cris, sub voce.
GRAMMATICAL STRUCTURE OF THE LENAPE. 105
Grammatical Structure of the Lenape.
It would not be worth while for me to enter into the in-
tricacies of Lenape grammar, particularly as I can add little
to what is already known.
The Delaware Grammar of Zeisberger remains our only
authority, and in spite of its manifest shortcomings and state
of incompletion, the unprejudiced student must acknowledge,
with Albert Gallatin,^ that it is "most honestly done," and
showed the Delaware as it actually was spoken, though perhaps
not as scientific linguists think it ought to have been spoken.
A few general observations will be sufficient.
As in other languages of the class, the theme is indifferently
nominal, verbal or adjectival \ that is, it performs the functions
of either of these grammatical categories, according to its
connection.
Nominal themes are either animate or inanimate. The
characteristic of all animate plurals is k {ak, ik, ek). Inani-
mate plurals are in al, wall or a. As usual, the distinction
between animate and inanimate nouns is partly logical, partly
grammatical, various objects being conceived as animate which
are in fact not so.
The possessive relation is generally indicated by placement
^ In Trans. Afner. Antiq. Society, Vol. II, p. 223. Zeisberger's state-
ments were criticised by Joseph Howse, Grammar of the Cree Language,
pp. 109, 310, 313. His strictures and those of the Abb6 Cuoq, in his
Etudes Philologiques stir Qiielques Langues Sauvages, Chap. I, were col-
lected and extended by Dr. J. Hammond Trumbull, in his paper on
"Some Mistaken Notions of Algonquin Grammar," Trans, of the Ameri-
can Philological Association, 1874. There is a needless degree of severity
in both these last named productions.
H
106 THE LENAPE AND THEIR LEGENDS.
alone, the possessor preceding the thing possessed, as le7i7io
quisall, the man's son ; but one could also say leJtJio w' quisall,
the man his son.
Adjectives precede nouns, and when used attributively
assume a verbal form by adding the termination wi, which
indicates objective existence (like the Chip. -wiii). Thus,
scattek, burning ; scattetm w'dehln, a burning heart — literally,
it-is-a-burning-thing his-heart.
The degrees of comparison are formed by prefixing allo-
wiwi, more, and eluwi, most. Both of these are from the
same radical aid, which may perhaps come from the admira-
tionis particula, ala" (Abnaki, ara') found in the northern
dialects as expressive of astonishment.^
There being no relative pronoun in Delaware, dependent
clauses are either included in the verbal of the major clause,
or include it as a secondary.
The scheme of the simple sentence is usually subject-verb-
object ; but emphasis allows departures from this, as in the
following sentence from Bishop Ettwein's MSS. : —
Jesus wemi amemensall w' t alio law ak.
Jesus all children he-loved-them.
Of the formal affixes, the inseparable pronouns are the
most prominent. They are the same for nouns and verbs,
and are —
ist. n, I, my, we, our.
2d. k, thou, thy, you, your.
3d. w or 0, he, she, it, his, their.
^ Rasles, Dictionary of the Abnaki, p. 550. Dr. Trumbull compares
the Mass. aniie, more than. Trans. American Philological Association,
1872, p. 168.
GRAMMATICAL STRUCTURE OF THE LENAPE. 107
Past time is indicated by the terminal/, with a connective
vowel, and future time by tsch, which may be either a prefix
or suffix, as —
N'' dellsin, I am thus.
N' dellsineep, I was thus.
N ' dellsintschi, ")
or . ^ I shall be thus.
Nantsch n dellsin, )
The change or "flattening" of the vowel of the root in
suppositive propositions, was recognized as a fact of speech,
but not grammatically analyzed by Zeisberger,
Its effect on verbal forms may be seen from the following
examples from his Grammar : —
Examples of Vowel Change in Lenape.
N'dappln, I am there. Achpiya, if I am there.
Epia, where I am.
N' dellsin, I am so. Lissiye, if I am so.
N'gauwi, I sleep. Gewi, he who sleeps.
N'pommauchsi, I walk or live. Pemauchsit, living.
N'da, I go. Eyaya, when I go.
Eyat, going.
Another omission in his Grammar is that of the ''obvia-
tive" and " super-obviative" forms of nouns. These are
used in the Algonkin dialects to define the relations of third
persons. They prevent such obscurity as appears in the
following English sentence : "John's brother called at Rob-
ert's, to see his wife." Whose wife is referred to is left
ambiguous ; but in Algonkin these third persons would have
different forms, and there would be no room for ambiguity.
In his writings in Lenape, Zeisberger makes use of obviatives,
108 THE LENAPE AND THEIR LEGENDS.
with the terminations al and I, but does not treat of them in
his Grammar.
As a question in philosophical grammar, it may be doubted
whether the Lenape has any true passive voice. Cardinal
Mezzofanti was accustomed to deny the presence of any real
passives in American languages; and he had studied the
Delaware among others.
The sign of the Delaware passive is the suffix gussu or cusso.
In the Cree dialect, which, as I have already said, preserves
the ancient forms most closely, this is k-ussu, and is a particle
expressing likeness or similarity in animate objects.^ Hence,
probably, the original sense of the Lenape word translated,
"I am loved," is "lam like the object of the action of
loving."
^ J. Howse, Grammar of the Cree Language, p. iii.
CHAPTER V.
Historical Sketches of the Lenape.
§ I. The Lenape as " Women."
§ 2, Recent Migrations of the Lenape.
g 3. Missionary Efforts in the Provinces of Pennsylvania and New Jersey.
The Lenape as '^ Women.''''
A unique peculiarity of the political condition of the
Lenape was that for a certain time they occupied a recog-
nized position as non-combatants — as ''women," as they
were called by the Iroquois.
Indian customs and phraseology attached a two-fold sig-
nificance to this term.
The more honorable was that of peace-makers. Among
the Five Nations and Susquehannocks, certain grave matrons
of the tribe had the right to sit in the councils, and, among
other privileges, had that of proposing a cessation of hostili-
ties in time of war. A proposition from them to drop the
war club could be entertained without compromising the
reputation of the tribe for bravery. There was an official
orator and messenger, whose appointed duty it was to convey
such a pacific message from the matrons, and to negotiate
for peace. ^
Another and less honorable sense of the term arose from a
custom prevalent throughout America, and known also among
the ancient Scythians. Its precise purpose remains obscure,
^ H. R. Schoolcraft, Notes on the Iroquois^ pp. 135-36.
109
110 THE LENAPE AND THEIR LEGENDS.
although it has been made the subject of a careful study by
one of our most eminent surgeons, who had facilities of
observation among the Western tribes/ Certain young men
of the tribe, apparently vigorous and of normal development,
were deprived of the accoutrements of the male sex, clothed
like women, and assigned women's work to do. They neither
went out to hunt nor on the war-path, and were treated as
inferiors by their male associates. Whether this degradation
arose from superstitious rites or sodomitic practices, it cer-
tainly carried to its victims the contempt of both sexes.
In their account of the transaction the Delawares claimed
that they were appointed as peace-makers in an honorable
manner, although the Iroquois deceived them as to their
object.
The Lenape account is as follows : —
"The Iroquois sent messengers to the Delawares with the
following speech : —
" ' It is not well that all nations should war; for that will
finally bring about the destruction of the Indians. We have
thought of a means to prevent this before it is too late. Let
one nation be The Woman. We will place her in the middle,
and the war nations shall be the Men and dwell around the
Woman. No one shall harm the Woman ; and if one does,
we shall speak to him and say, 'Why strikest thou the
Woman ? ' Then all the Men shall attack him who has
1 The Disease of the Scythians (^Morbus Feminariun') and Certain
Analogous Conditions. By William A. Hammond, M. D. (New York, 1882).
Dr. Hammond found that the hombre miijerado of the Pueblo Indians " is
the chief passive agent in the pederastic ceremonies which form so im-
portant a part in their religious performances," p. 9.
THE LENAPE AS "WOMEN." Ill
Struck the Woman. The Woman shall not go to war, but
shall do her best to keep the peace. When the Men around
her fight one another, and the strife waxes hot, the Woman
shall have power to say : ' Ye Men ! what do ye that ye thus
strike one another ? Remember that your wives and children
must perish, if ye do not cease. Will ye perish from the face
of the earth ? ' Then the Men shall listen to the Woman and
obey her.'
" The Delawares did not at once perceive the aim of the
Iroquois, and were pleased to take this position of the Woman.
''Then the Iroquois made a great feast, and invited the
Delawares, and spoke to their envoys an address in three
parts.
" First, they declared the Delaware nation to be the Woman
in these words: —
" ' We place upon you the long gown of a woman, and
adorn you with earrings.'
" This was as much as to say that thenceforward they were
not to bear arms.
" The second sentence was in these words : —
" ' We hang on your arm a calabash of oil and medicine.
With the oil you shall cleanse the ears of other nations that
they listen to good and not to evil. The medicine you shall
use for those nations who have been foolish, that they may
return to their senses, and turn their hearts to peace.'
" The third sentence intimated that the Delawares should
make agriculture their chief occupation. It was : —
"'We give herewith into your hands a corn pestle and a
hoe.'
" Each sentence was accompanied with a belt of wampum.
112 THE LENAPE AND THEIR LEGENDS.
These belts have ever since been carefully preserved and their
meanings from time to time recalled." ^
Opinions of historians about this tradition have been vari-
ous. It has generally been considered a fabrication of the
Delawares, to explain their subjection in a manner consoling
to their national vanity. Gen. Harrison dismisses it as
impossible;^ Albert Gallatin says, "it is too incredible to
require serious discussion ; ^ Mr. Hale characterizes it as
"preposterous;"* and Bishop de Schweinitz as "fabulous
and absurd." ^
On the other hand, it is vouched for by Zeisberger, who
furnished the account to Loskiel, and who would not have
said that the wampum belts with their meaning were still pre-
served unless he knew it to be a fact. It is repeated em-
phatically by Heckewelder, who adds that his informants
were not only Delawares but Mohegans as well, who could not
have shared the motive suggested above.*
There can be no question but that the neutral position of
the Delawares was something different from that of a con-
quered nation, and that it meant a great deal more. They
undoubtedly were the acknowledged peace-makers over a
wide area, and this in consequence of some formal ancient
^ Loskiel, GescJiichte der Mission, etc., s. i6i-2.
* Wm. Henry Harrison, A Discourse on the Aborigines of the Valley of
the Ohio, pp. 24, 25 (Cincinnati, 1838).
5 Gallatin, Trans. Amer. Antiq. Soc, Vol. II, p. 46.
* Horatio Hale, The Iroquois Book of Rites, p. 92.
Edmund de Schweinitz, Life and Times of David Zeisberger, p. 46.
Heckewelder, Indian Nations, pp. xxxii and 60.
THE LENAPE AS "WOMEN." 113
treaty. This is distinctly stated by the Stockbridge Indian,
Hendrick Aupaumut, in his curious Narrative: — ^
''The Delawares, who we calld Wenaumeeti, are our Grand-
fathers, according to the ancient covenant of their and our
ancestors, to which we adhere without any deviation in these
near 200 years, to which nation the 5 nations and British have
commit the whole business. For this nation has the greatest
influence with the southern, western and northern nations."
Hence Aupaumut undertook his embassy directly to them,
so as to secure their influence for peace in 1791.
To the fact that they exerted this influence during the
Revolutionary War, may very plausibly be attributed the suc-
cess of the Federal cause in the dark days of 1777 and 1778 ;
for, as David Zeisberger wrote : " If the Delawares had taken
part against the Americans in the present war, America would
have had terrible experiences ; for the neutrality of the Dela-
wares kept all the many nations that are their grandchildren
neutral also, except the Shawanese, who are no longer in
close union with their grandfathers."^
^ Narrative of Hendrick Aupaumut, Mettts. Hist. Soc. Pa., Vol. II,
pp. 76-77. Wenaumeen for Unami, the Mohegan form of the name.
This seems to limit the peace-making power to that gens. He may mean,
" Those of the Delawares who are called the Unamis are our Grand -
fathers," etc.
The Chipeways, Ottawas, Shawnees, Pottawattomies, Sacs, Foxes and
Kikapoos, all called the Delawares " Grandfather." J. Morse, Report
on Indian Affairs, pp. 122, 123, 142. The term was not intended in a
genealogical, but solely in a political, sense. Its origin and precise mean-
ing are alike obscure.
2 History of the Indians, MS., quoted by Bishop Schweinitz, life of
Zeisberger, p. 444, note.
114 THE LENAPE AND THEIR LEGENDS.
When at the close of the French War, in 1758, the treaty
of Easton put a stop to the bloody feuds of the border, "the
peace-belt was sent to our brethren, the Delawares, that they
might send it to all the nations living toward the setting
sun,"^ and they carried it as the recognized pacific envoys.
The Iroquois, however, assumed a most arrogant and con-
temptuous tone toward the Delawares, about the middle of
the eighteenth century. In 1756 they sent a belt to them,
with a most insulting message:^ "You will remember that
you are our women ; our forefathers made you so, and put a
petticoat on you, and charged you to be true to us, and lie
with no other man ; but now you have become a common
bawd," etc.
Two years later, the Cayuga chief, John Hudson, said, at
a council at Burlington,^ "The Munseys are women, and
cannot make treaties for themselves."
These were but repetitions of the famous diatribe of the
Onondaga chieftain, Canassatego, at a council at Philadel-
phia, in 1742. Turning to the representatives of the Lenape,
he broke out upon them with the words : —
" How came you to take upon you to sell land ? We con-
quered you. We made women of you. You know you are
women, and can no more sell land than women, * * *
We charge you to remove instantly. We don't give you the
liberty to think about it. We assign you two places to go
1 The words are those of George Croghan, Esq., at the treaty of Pitts-
burg, 1759, with the Six Nations and Wyandots. History of Western
Penna., App. p. 135.
2 Records of the Council at Easton, 1756, in Lib. Amer. Philos. Soc.
s Smith, History of New Jersey, p. 451 (2d ed.)
THE LENAPE AS "WOMEN." 115
to, either Wyoming or Shamokin. Don't deliberate, but
remove away; and take this belt of wampum."
And as he handed the belt to the Lenape head chief he
seized him by his long hair and pushed him out of the door
of the council room !
It was notorious at the time, however, that this was a scene
arranged between the Governor of the Province, Mr. George
Thomas, and the Iroquois deputation. The Lenape had been
grossly cheated out of their lands by the trick of the so-called
"Long Walk," in 1735, and they refused to vacate their
hunting grounds. The Governor sent secret messengers to
the pov/erful and dreaded Six Nations to exert their pretended
rights, and paid them well for it.^
What could the Lenape do ? They were feeble, and un-
doubtedly had been brought under the authority of their
warlike northern neighbors. They found themselves in the
position of the Persian chieftain Harmosar, as he stood before
the caliph Omar, and heard the latter revile the patriot cause :
" In deinen Handen ist die Macht,
Wer einem Sieger widerspricht, der widerspricht mit Unbedacht."
— Von Platen- Hallermunde.
Such were the respective claims of the Lenape and Iroquois.
Instead of discussing the antecedent probability of one or the
other being true, I shall endeavor to ascertain from the early
records the precise facts about this curious transaction.
1 See the Narrative of the Long Walk, by John Watson, father and son,
in Hazard's Register of Fenna., 1830, reprinted in Beach's Indian Mis-
cellany, pp. 90-94; also the able discussion of the question in Dr. Charles
Thompson's Inquiry into the Causes of the Alienation of the Delaware and
Shawnee Indians, pp. 30—34 and 42-46. (London, 1759.)
116 THE LENAPE AND THEIR LEGENDS.
It is certain that toward the close of the sixteenth century
the unending wars between the Delaware confederacy and the
Iroquois had reduced the latter almost to destruction. The
Jesuit missionaries tell us this.^ The turning point in their
affairs was the settlement of the Dutch on the Hudson.
Quick to appreciate the value of firearms, they bought guns
and powder at any price, and soon had rendered themselves
formidable to all their neighbors.^ About 1670 they
attacked successfully that family of the Minsi called the
Minisink.
This was probably the victory to which the Five Nations
referred at a treaty at Philadelphia, in 1727, when they stated
that their conquest of the Delawares was about the time
William Penn first landed, and that he sent congratulations
to them on their success — an obvious falsehood.^
They were certainly at that period pressing hard on the
Susquehannocks and destroying their remnant in the valley
of that river. Mr. William P. Foulke is quite correct in his
conclusion that, " Upon the whole we may conclude that the
^ Relations des Jesuiies, 1 660, p. 6. Some confusion has arisen in this
matter, from confounding the Susquehannocks with the Iroquois, both of
whom were called "Mengwe" by the Delawares, corrupted into " Min-
goes." Thus, a writer in the first half of the 17th century says of the
" Mingoes" that the river tribes " are afraid of them, so that they dare not
stir, much less go to war against them." Thomas Campanius, Description
of the Province of Nezv Siveden, p. 158.
2 See Mr. E. M. Ruttenber's able discussion of the subject in his
History of the Indian Tribes of Hudsott's River, p. 66 (Albany, 1872).
* Dr. Charles Thompson, An Inquiry iiito the Causes of the Alienation
of the Delaware aiid Shawnee Indians, pp. 11, 12. (London, 1759.)
THE LENAPE AS "WOMEN." 117
Lancaster lands fell into the power of the Five Nations at
some time between 1677 and 1684."^
Yet their conquest of the Minsi was not complete. The
latter had the mind and the will to renew the combat. In
1692 they appealed to the government of Pennsylvania to aid
them in an attack on the Senecas, but the Quakers declined
the foray. The next year the Minsi asked Governor Benjamin
Fletcher at least to protect them against these Senecas, adding
that with assistance they were ready to attack them, for
" although wee are a small number of Indians, wee are Men,
and know fighting,"^
Evidently there was neither subjection nor womanhood
with the Minsi at that date.
There is also positive evidence that the Five Nations at
that time regarded the Delawares as a combatant nation, and
worthy of an invitation to join a war. On July 6th, 1694,
Governor Wm. Markham met in conference the famous chief
Tamany and others; and the Delaware orator, Hithquoquean,
laid down a belt of wampum, and said : — ^
" This belt is sent us by the Onondagas and Senecas, who
say : ' You Delaware Indians do nothing but stay at home
and boil your pots, and are like women ; while we, Onondagas
and Senecas, go abroad and fight the enemy.'
'' The Senecas would have us Delaware Indians to be part-
ners with them, and fight against the French, but we, having
1 See his " Notes Respecting the Indians of Lancaster County, Penna.,"
in the Collections of the Historical Society of Penna., Vol. IV, Part
p. 198.
2 Minutes of the Provincial Council of Pennsylvania, Vol. I, p. 333.
» Ibid, Vol. I, p. 410-11.
118 THE LENAPE AND THEIR LEGENDS.
always been a peaceful people, and resolving to live so ; and
being but weak and verie few in number, cannot assist them,
and having resolved among ourselves not to go, doe intend
to send back, this their Belt of Wampum."
The Lenape, therefore, did not, at that date, occupy any
degrading position, although they were under the general
domination of the Iroquois League.
Both these points are proved yet more conclusively by the
proceedings at a conference at White Marsh, May 19th, 1712,
between Governor C. Gookin and the Delaware chiefs.
Gollitchy, orator of the latter, exhibited thirty-two belts of
wampum, which they were on their way to deliver to the
Five Nations, adding "that many years ago they had been
made tributaries to the Mingoes." He also shewed "a long
Indian pipe, with a stone head, a wooden shaft, and feathers
fixt to it like wings. This pipe, they said, upon making
their submission to the Five Nations, who had subdued
them, and obliged them to be their tributaries, those Nations
had given to these Indians, to be kept by them." All the
tribute belts, however, were sent by the women and chil-
dren, as the speaker explained at length, "as the Indian
reckons the paying of tribute becomes none but women and
children. "1
Fortunately, however, we are able to fix the exact date
and circumstances of the political transformation of the
Delawares into women. It is by no means so remote as
Mr. Heckewelder thought, who located the occurrence at
Norman's Kill, on the Hudson, between 1609 and 1620;^
1 Minutes of the Provincial Council, Vol. II, pp. ST^-73-
2 History of the Indian Nations, p. xxix.
THE LENAPE AS "WOMEN." 119
and it was long after 1670, which is the date assigned by
Mr. Ruttenber/ from a study of the New York records.
It was in the year 1725, and was in consequence of the
Delawares refusing to join the Iroquois in an attack on the
English settlements.
These data come to light in a message of the Shawnee
chiefs, in 1732, to Governor Gordon, who had inquired their
reasons for migrating to the Ohio Valley.
Their reply was as follows : —
"About nine years agoe the 5 nations told us att Shallyschoh-
king, wee Did nott Do well to Setle there, for there was a Greatt
noise In the Greatt house and thatt in three years time, all Should
know whatt they had to Say, as far as there was any Setlements
or the Sun Sett.
" About ye Expiration of 3 years afifore S'^, the 5 nations Came
and Said our Land is goeing to bee taken from us. Come brothers
assistt us Lett us fall upon and fightt with the English. Wee
answered them no, wee Came here for peace and have Leave to
Setle here, and wee are In League with them and Canott break itt.
" Aboutt a year after they, ye 5 nations. Told the Delawares
and us, Since you have nott hearkened to us, nor Regarded whatt
we have said, now wee will pettycoatts on you, and Look upon
you as women for the future, and nott as men. Therefore, you
Shawanese Look back toward Ohioh, The place from whence you
Came, and Return thitherward, for now wee Shall Take pitty on
the English and Lett them have all this Land.
"And further Said now Since you are Become women, He
Take Peahohquelloman, and putt itt on Meheahoaming and He
Take Meheahoaming and putt itt on Ohioh, and Ohioh He putt
on Woabach, and thatt shall bee the warriours Road for the
future." {Petma. Archives, Vol. I.)
1 The Indian Tribes of Hudson's River, p. 69.
120 THE LENAPE AND THEIR LEGENDS.
The circumstances attending the ceremony were probably
pretty much as Loskiel relates.
The correctness of this account is borne out by an ex-
amination of law titles.
That the river tribes at the time of Penn's treaties (1680-
1700) could not sell their lands without the permission of the
Iroquois has never been established. Mr, Gallatin states that
William Penn " always purchased the right of possession from
the Delawares, and that of sovereignty from the Five Nations."^
This may have been the case in some later treaties of the
colony, but certainly there is no intimation of it in the cele-
brated "First Indian Deed" to Penn, July 15th, 1682.^
Furthermore, in the Release which the Iroquois did give of
their Pennsylvania lands in 1736, the boundaries are defined
as " Westward to the Setting of the Sun, and Eastward to the
furthest springs of the Waters running into the said River,"
/. e., the Susquehannah ; '^ and to do away with any doubt
that the tract thus defined included all the land in this part
to which they had a claim, the Release goes on to recite that
"our true intent and meaning was and is to release all our
Right, Claim and Pretensions whatsoever to all and every the
Lands lying within the Bounds and Limits of the Government
of Pennsylvania, Beginning Eastward on the River Delaware,
as far Northward as the s* Ridge or Chain of Endless Moun-
tains." In other words, although the Six Nations advanced'
no claim to land east of the Susquehanna watershed, the
Proprietors chose to include the Delaware watershed so as to
1 Trans. Am. Antiq. Soc, Vol. II, p. 46.
2 Pennsylvajiia Archives, Vol. I, p. 47.
* Pennsylvania Archives, Vol. I, p. 498.
THE LENAPE AS "WOMEN." 121
avoid any future complication. It seems to me this Release
does away with any " right of sovereignty " of the Iroquois
over the Delaware Valley south of the mountains, and brands
Canassatego's remarks above quoted as braggart falsehoods.
As for land east of the Delaware river, Mr. Ruttenber cor-
rectly observes : ''The Iroquois never questioned the sales
made by the Lenapes or Minsis east of that river. * * The
findings of Gallatin in this particular are confirmed by all the
title deeds in New York and New Jersey." ^
It was only to the Susquehannock lands, purchased by Penn
in 1699, that the confirmation of the Iroquois was required.*
The close of this condition of subjection was in 1756. In
that year Sir William Johnson formally ''took off the petti-
coat " from the Lenape, and " handed them the war belt."^
The year subsequent they made the public declaration that
" they would not acknowledge but the Senecas as their su-
periors." *
Even their supremacy was soon rejected. At the Treaty of
Fort Pitt, October, 1778, Captain White Eyes, when reminded
by the Senecas that the petticoats were still on his people,
scornfully repudiated the imputation, and made good his
words by leading a war party against them the following
year.
^ The Indian Tribes of Hudson' s River, p. 69.
* See Penna. Archives^ Vol. I, p. 144, and Du Ponceau, Memoir on the
Treaty at Shackatnaxon, Collections of the Penna. Hist. Soc, Vol. Ill,
Part II, p. 73.
* New York Colonial Documents, Vol. VII, p. 119.
4 Thompson, Inquiry into the Causes of the Alienation of the Delaware
and Shawnee Indians, p. 107.
I
122 THE LENAPE AND THEIR LEGENDS.
The Iroquois, however, released their hold unwillingly, and
it was not until 1794, shortly before the Treaty of Greenville,
that their delegates came forward and " officially declared
that the Lenape were no longer women, but men,'''' and the
famous chief, Joseph Brant, placed in their hands the war
club/
§ 2. Histoi'ic Migrations of the Lenape.
It does not form part of my plan to detail the later history
of the Lenape. But some account of their number and mi-
grations will aid in the examination of the origin and claims
of the Walum Olum,
The first estimate of the whole number of native inhabitants
of the province was by William Penn. He stated that there
were ten different nations, with a total population of about
6000 souls. ^
This was in 1683. Very soon after this they began to
diminish by disease and migration. As early as 1690, a band
of the Minsi left for the far West, to unite with the Ottawas.''
In 1 721 the Frenchman Durant speaks of them as "exceed-
ingly decreased."* Already they had yielded to the pressure
of the whites, and were seeking homes on the head-waters of
the Ohio, in Western Pennsylvania. Their first cabins are
said to have been built there in 1724.^
1 Heckewelder, Indian Nations, p. 70; E. de Schweinitz, Life of
Zeisberger, pp. 430, 641.
2 Janney, Life of Penn, p. 247.
3 Ruttenber, Indians of the Hudson Rive?-, p. 177.
4 Durant's Afe??torial, in New York Colonial Documents, Vol. V, p. 623.
5 Early History of Western Pennsylvania, p. 31 (Pittsburg, 1846);
and see Penna, Archives, Vol. I, pp. 322, 330.
HISTORIC MIGRATIONS OF THE LENAPE. 123
All that remained in the Delaware valley were ordered by
the Iroquois, at the treaty of Lancaster, 1744, to leave the
waters of their river, and remove to Shamokin (now Sipbury)
and Wyoming, on the Susquehanna, and most of them
obeyed. The former was their chief town, and the residence
of their "king," Allemoebi.
When the interpreter, Conrad Weiser, visited their Ohio
settlements, in 1748, he reported their warriors there at 165,
which was probably about one-fourth of the nation.
In the "French War," 1755, the Delawares united with
the French against the Iroquois and English, and suffered
considerable losses. At its close they were estimated to
have, both on the Susquehanna and in Ohio, a total of 600
available fighting men.^
After this date they steadily migrated from the Susque-
hannah to the streams in central and eastern Ohio, es-
tablishing their chief fire on the Tuscarawas river, at
Gekelemukpechunk, and hunting on the Muskingum, the
Licking, etc.^
When the war of the Revolution broke out, Zeisberger
used all his efforts to have them remain neutral, and at least
prevented them from joining in a general attack on the settle-
ments. Their distinguished war-chief, Koquethagachton,
known to the settlers as "Captain White Eyes," declared,
in 1775, in favor of the Federal cause, and renounced for
himself and his people all dependence on the Iroquois.
These friendly relations were confirmed at the treaty of
^ Loskiel, Geschichte der Missiott, p. 54. The treaty of Lancaster
1762, was the last treaty held with the Indians in eastern Pennsylvania.
* Schweinitz, Life of Zeisberger, p. 90.
124 THE LENAPE AND THEIR LEGENDS.
Fort Pitt (177S), and the next year a number of Delawares
accompanied Col. Brodhead in an expedition against the
Senecas.
The massacre of the unoffending Christian natives of
Gnadenhiitten, in 1788, was but one event in the murderous
war between the races that continued in Ohio from 1782 to
the treaty of peace at Greenville, in 1795.
To escape its direful scenes, a part of the Delawares re-
moved south, to upper Louisiana, in 1 789, where they received
official permission from Governor Carondelet, in 1793, to
locate permanent homes. ^ Zeisberger also, in 1791, conducted
his colony of Christian Indians to Canada, and founded the
town of Fairfield, on the Retrenche river. Thus, in both
directions the Delawares were driven off the soil of the
United States. Yet those that remained in Ohio, if we
may accept the account of John Brickell, who was a captive
among them from 1791 to 1796, attempted to live a peaceable
and agricultural life.^
Peace restored, the Delawares made their next remove to
the valley of White Water river, Indiana, where they attempted
to rekindle the national council fire, under the head chief
Tedpachxit. They founded six towns, the largest of which
was Woapikamikunk or Wapejninskink, "Place of Chestnut
Trees." This tract was guaranteed them "in perpetuity"
^ New York Colonial Documents^ Vol. VH, p. 583.
2 On the locations of the Delawares in Ohio, and the boundaries of their
tract, see Ed. de Schweinitz, Life of Zeisberge?; p. 374, and an article by
the Rev. Stephen D. Peet, entitled " The Delaware Indians in Ohio," in
the American Antiquarian, Vol. II.
HISTORIC MIGRATIONS OF THE LENAPE. 125
by the treaty of Vincennes, 1808.^ Nevertheless, just ten
years later, at the treaty of St. Mary's, they released the
whole of their land, ''without reserve," to the United States,
the government agreeing to remove them west of the Missis-
sippi, and grant them land there.
At this time they numbered about 1000 souls, of whom
800 were Delawares, the others being Mohegans and Nanti-
cokes.^ Their head chief was Thahutoowelent, of the Turkey
tribe, Tedpachxit having been assassinated, at the instigation
of Tecumseh.
They are described as ''having a peculiar aversion to white
people," and "more opposed to the Gospel and the whites
than any other Indians,"^ which is small matter of wonder,
when they had seen the peaceful Christian converts of their
nation massacred three times, in cold blood, once at Gnad-
enhiitten, in Pennsylvania (1756); again at Gnadenhiitten,
in Ohio (1788), and finally at Fairfield, Canada (1813).
The Rev. Isaac McCoy, who visited them on the White
Water, in the winter of 181 8-1 9, states that they lived in
log huts and bark shanties, and were fearfully deteriorated
by whisky drinking.*
The last band of the Delawares that appeared in Ohio was
in 1822.^
1 The position of the Delawares in Indiana is roughly shown on
Hough's Map of the Tribal Districts of Indiana, in the Report on the
Geology and Natural History of Indiana, 1882.
2 J. Morse, Report on the Indian Tribes, p. no.
3 Mr. John Johnston, Indian Agent, in Trans, of the Amer. Anti-
quarian Society, Vol. I, p. 271.
* History of the Baptist Indian Missions, p. 53, etc.
5 Captivity of Christian Fast, in Beach, Indian Miscellajiy, p. 63.
126 THE LENAPE AND THEIR LEGENDS.
The location assigned to the Delawares was near the mouth
of the Kansas river, Kansas. They were reported, in 1850,
as possessing there 375,000 acres and numbering about 1500
souls. Four years later they "ceded" this land, and were
moved to various reservations in the Indian Territory.
There still remain about sixty natives at New Westfield,
near Ottawa, Kansas, under the charge of the Moravian
Church. The same denomination has about 300 of the tribe
on the reservation at Moraviantown, in the province of
Ontario, Canada. A second reservation in Canada is under
the charge of the Anglican Church. The majority of the tribe
are scattered in different agencies in the Indian Territory.
§ 3. Missionary Efforts in the Provinces of New Jersey
and PeJinsylvania.
None of the American colonies enjoyed a more favorable
opportunity to introduce the Christian religion to the natives
than that located on the Delaware river. What use was
made of it ?
The Rev. Thomas Campanius, of Stockholm, a Lutheran
clergyman, attached to the Swedish settlement from 1642 to
1649, made a creditable effort to acquire the native tongue and
preach Christianity to the savages about him. He translated
the Catechism into the traders' dialect of Lenape, but we have
no record that he succeeded in his attempts at conversion.
One might suppose that so very religious a body as the
early Friends would have taken some positive steps in this
direction. Such was not the case, I have not found the record
of any one of them who set seriously to work to learn the native
tongue, without which all effort would have been fruitless.
MISSIONARY EFFORTS. 127
William Penn was not wholly unmindful of the spiritual
condition of his native wards. In 1699 he offered to provide
the Friends' Meeting at Philadelphia with interpreters to con-
vey religious instruction to the Indians. But the Meeting
took no steps in this direction. He himself, when in the
colony in 1701, made some attempts to address them on re-
ligious subjects, as did also Friend John Richardson, who was
with him, availing themselves of interpreters. The latter
reports a satisfactory response to his words, but not being
followed up, their effect was ephemeral.^
Nothing further was done for nearly half a century, and
when the enthusiastic young David Brainerd began his mission
in 1742, he distinctly states that there was not another mis-
sionary in either province.^ His labors extended over four
years, and were productive of some permanent good results
among the New Jersey Indians, and this in spite of the sus-
picions, opposition and evil example of the whites around
him. The little society of Christian Indians which he gathered
in Burlington County, New Jersey, was even reported as a
congregation of rioters and enemies of the State ! ^
1 See the work entitled, Account of the Conduct of the Society of Friends
toward the Indian Tribes, pp. 55 seq. (London, 1844.)
2 " I have likewise been wholly alone in my work, there being no other
missionary among the Indians, in either of these Provinces." He wrote
this in 1746. Life of David Brainerd, p. 409.
* See " A State of Facts about the Riots," in New Jersey Archives,
Vol. VI, pp. 406-7, where the writer speaks with great suspicion of " the
cause pretended for such a number of Indians coming to live there is that
they are to be taught the Christian religion by one Mr. Braniard." Well
he might ! Any such occurrence was totally unprecedented in the annals
of the colony.
128 THE LENAPE AND THEIR LEGENDS.
Nor was the province of Penn inclined to greater favors
toward Christianized natives. When the Indians were cheated
out of their lands by the " Long Walk," a few who had been
converted, among others the chief Moses Tatemy, petitioned
the Council to remain on their lands, some of which were
direct personal gifts from the Proprietaries. Their request
was refused, and Moses Tatemy, who did remain, was shot
down like a dog, in the road, by a white man.^
Unknown to Brainerd, however, the seeds of a Christian
harvest had already been sown, in 1742, in the wilderness of
Pennsylvania, by the ardent Moravian leader. Count Nicholas
Lewis Zinzendorf; already, in 1744, the fervent Zeisberger,
prescient of his long and marvelous service in the church
militant, had registered himself as destinirter Heidenbote —
"appointed messenger to the heathen" — in the corner-stone
of the Brethren's House, at Bethlehem; already the pious
Rauch had collected a small but earnest congregation of
Mohegans at Shekomeko, who soon removed to the Lehigh
valley, and pitched the first of those five Gnadenhiltten,
"Tents of Grace," destined successively to mark the un-
wearied efforts of the Moravian missionaries, and their frus-
tration through the treachery of the conquering whites.^
1 See Minutes of the Provincial Council of Fenna., Nov., 1742, Vol.
IV, 624-5. Further, on Tatemy, who had been converted by Brainerd
and served him as interpreter, see Heckevifelder, Indian Nations, second
edition, p. 302, note of the editor.
2 The Heckewelder MSS., in the library of the Am. Philos. Society, give
the results of the first twenty years, 1741-61, of the labors of the Moravian
brethren. In that period 525 Indians were converted and baptized. Of
these — 163 were Connecticut Wampanos; 11 1 were Mahicanni proper;
251 were Lenape. Some of the latter were of the New Jersey Wapings.
MISSIONARY EFFORTS. 129
It is not my purpose to tell the story of this long struggle.
Its thrilling events are recounted, with all desirable fullness,
in the vivid narrative of Bishop Edmund de Schweinitz,
grouped around the marked individuality of the devoted
Zeisberger — pages which none can read without amazement
at the undaunted courage of these Christian heroes, without
sorrow at the sparse harvest gleaned from such devotion.^
When, after sixty-two years of missionary labors, the ven-
erable Zeisberger closed his eyes in death (1808), the huts of
barely a score of converted Indians clustered around his little
chapel. His aspiration that the Lenape would form a native
Christian State, their ancient supremacy revived and applied
to the dissemination of peace, piety and civilization among
their fellow-tribes — this cherished hope of his life had forever
disappeared. He had lived to see the Lenape, a mere broken
remnant, "steeped in all the abominations of heathenism,
eke out their existence far away from their former council
fires."
^ The Life and Times of David Zeisberger, the Western Pioneer and
Apostle of the Indians. By Edmund de Schweinitz, Philadelphia, 1871.
CHAPTER VL
Myths and Traditions of the Lenape.
Cosmogonical and Culture Myths.— The Culture-hero, Michabo.— Myths from Lind-
strom, Ettwein, Jasper Donkers, Zeisberger. — Native Symbolism. — The Saturnian
Age. — Mohegan Cosmogony and Migration Myth.
National Traditions. — Beatty's Account. — The Number Seven. — Heckewelder's Ac-
count. — Prehistoric Migrations. — Shawnee Legend. — Lenape Legend of the Naked
Bear.
Cosmogonical and Culture Myths.
The Algonkins, as a stock, had a well developed creation-
myth and a culture legend, found in more or less completeness
in all their branches.
Their culture hero, their ancestor and creator, he who made
the earth and stocked it with animals, who taught them the
arts of war and the chase, and gave them the Indian corn,
beans and squashes, was generally called Michabo, The Great
Light, but was also known among the Narragansetts of New
England as Wetucks, The Common Father; among the Cree
as Wisakketjdk, the Trickster ; by the Chippeways as Nana-
bozho (^JVendboj), the Cheat ; by the Black Feet as Natose,
Our Father, or Napiw ; and by the Micmacs and Penobscots
as Glus-Kap, the Liar.
I have given the details of this myth and analyzed them
in previous works ; ^ here it is sufficient to say that it is a
1 D. G. Brinton, Myths of the New World, Chap. VI. (N. Y., 1876), and
American Hero Myths, Chap. II (Phila., 1882). The seeming incongruity
of applying such terms as Trickster, Cheat and Liar to the highest divinity
I have explained in a paper in the American Antiqziarian for the current
year (1885) and will recur to later.
130
COSMOGONICAL AND CULTURE MYTHS. 131
Light-myth, and one of noble proportion and circumstance,
quite worthy of comparison with those of the Oriental
world.
Traces of it are reported among the Lenape, and I doubt
not that had we their ancient stories in their completeness,
we should find that they had preserved it as wholly as the
Chipeways. These related of their Nanabozho that he was
the son of a maiden who had descended from heaven. She
conceived without knowledge of man, and having given birth
to twins, she disappeared. One of these twins was Nanabozho.
Having formed the earth by his miraculous powers, and done
many wonderful things, he disappeared toward the east, where
he still dwells beyond the sunrise.
It was undoubtedly a fragment of this legend that the
Swedish engineer, Lindstrom heard among the Lenape, on
the Delaware, about 1650. They told him, or rather he
understood them, as follows : —
"Once, one of your women (z. e., a white woman) came
among us, and she became pregnant, in consequence of
drinking out of a creek ; an Indian had connection with her,
and she became pregnant, and brought forth a son, who,
when he came to a certain size, was so sensible and clever,
that there never was one who could be compared to him, so
much and so well he spoke, which excited great wonder ; he
also performed many miracles. When he was quite grown
up, he left us, and went up to heaven, and promised to come
again, but has never returned."^
This is but a mistranslation of the general Algonkin
legend, in which the virgin mother bears a white and dark
^ Thomas Campanius, Accotcni of New Sweden, Book III, cap. xi.
132 THE LENAPE AND THEIR LEGENDS.
twin, the former of whom becomes the tribal culture hero
and demiurgic deity.
Its interpretation is, that the virgin is the Dawn, who brings
forth the Day, which assures safety and knowledge, and the
Night, which departs with her. The Day leaves us, and in its
personified form returns no more, though ever expected.
That such were the original form and significance of the
myth, we have the testimony of Bishop Ettwein,^ himself a
Delaware scholar, and who drew his information from the
natives as well as the missionaries. He tells us that their
legend ran, that in the beginning the first woman fell from
heaven and bore twins ; that it was toward the east that they
directed their children to turn their faces when they prayed
to the spirits ; and that their old men had said that it was an
ancient belief that from that quarter some one would come
to them to benefit them. Therefore, said they, when our
ancestors saw the first white men, they looked upon them as
divine, and adored them.
The Dutch travelers, Jasper Donkers and Peter Sluyter,
relate a part of this myth as they heard it from New Jersey
Indians in 1679. These informed them that all things came
from a tortoise. It had brought forth the world, and from
the middle of its back had sprung up a tree, upon whose
branches men had grown.
This tortoise " had a power and a nature to produce all
things, such as earth, trees and the like." But it was not
t\\G primufii mobile, not the ultimate energy of the universe.
" The first and great beginning of all things was Kickeron
^ Traditions and Language of the Indians, in Bulletin Hist. Soc. Pa.,
Vol. I, pp. 30-31.
COSMOGONICAL AND CULTURE MYTHS. 133
or Kickerovi, who is the original of all, who has not only
once produced or made all things, but produces every day."
The tortoise brought forth what this primal divinity "wished
through it to produce."^
This is a very interesting statement. It reveals a depth of
thought on the part of the native philosophers for which we
were scarcely prepared. The worthy Dutch travelers do not
pretend to explain the myth. But its sense can be clearly
interpreted.
The turtle or tortoise is everywhere in Algonkin picto-
graphy the symbol of the earth.^ From the earth, from the
soil, all organic life, the whole realm of animate existence —
ever sharply defined in Algonkin grammar and thought from
inanimate existence — proceeds, directly as vegetable life, or
indirectly as animal life. The earth is the All-Mother, ever-
producing, inexhaustible.
As for Kikeron, the eternally active, hidden spirit of the
universe, I have but to refer the reader to the list of ideas
associated around this root kik, which I have given on a pre-
vious page (p. 102) to reveal the significance of this word.
We may, with equal correctness, translate it Life, Light,
Action or Energy. It is the abstract conception back of all
these.
The distinction was the same as that established by the
^ Joiirfial of a Voyage to JVew York in idyg-So. By Jasper Donkers
and Peter Sluyter, p. 268. Translation in Vol. I of the Transactions of
the Long Island Historical Society (Brooklyn, 1867).
2 Schoolcraft says of the Chipeway pictographic symbols : " The turtle
is believed to be, in all instances, a symbol of the earth, and is addressed
as mother." History and Statistics of the Indian Tribes, Vol. I, p. 390.
134 THE LENAPE AND THEIR LEGENDS.
scholastic philosophers between the mundus and the ^am'ma
mundi', between the essentia and the existentia; between
natura natiirans and natura naturata. But who expected to
find it among the Lenape ?
This creation myth of the Delawares is also given in brief
by Zeisberger. It dated back to that marvelous overflow
which is heard of in many mythologies. The whole earth
was submerged, and but a few persons survived. They had
taken refuge on the back of a turtle, who had reached so
great an age that his shell was mossy, like the bank of a
rivulet. In this forlorn condition a loon flew that way, which
they asked to dive and bring up land. He complied, but
found no bottom. Then he flew far away, and returned with
a small quantity of earth in his bill. Guided by him, the
turtle swam to the place, where a spot of dry land was found.
There the survivors settled and repeopled the land.^
This is more a tale of reconstruction than a creation myth.
It is that which has generally been supposed to refer to the
Deluge. But, as I have explained in my " Myths of the New
World," all these so-called Deluge Myths are but develop-
ments of crude cosmogonical theories.
To understand the significance of this myth we must
examine the Indian notion of the earth. This is the more
germane to my theme, as the meaning of the original text
which is printed in this volume can only be grasped by one
acquainted with this notion.
The Indians almost universally believed the dry land they
knew to be a part of a great island, everywhere surrounded
1 Zeisberger, MSS., in E. de Schweinitz, Life and Times of Zeisberger,
pp. 218, 219; Heckewelder, Indian Nations, p. 253.
COSMOGONICAL AND CULTURE MYTHS. 135
by wide waters whose limits were unknown/ Many tribes
had vague myths of a journey from beyond this sea; many
placed beyond it the home of the Sun and of Light, and the
happy hunting grounds of the departed souls. The Delawares
believed that the whole was supported by a fabled turtle, whose
movements caused earthquakes and who had been their first
preserver.^ As above mentioned, the turtle in its amphibious
character and rounded back represented the earth or the land
itself, as distinguished from water. Like the turtle, the land
lies at times under the water and at times above it. The
spirit of the earth was the practical and visible developmental
energy of nature.
The medicine men, or conjurers, who professed to be in
personal relations with this power, made their "medicine
rattle" of a turtle shell (Loskiel), and when they died,
such a shell was suspended from their tomb posts (Zeis-
berger).
The Delawares also shared the belief, common to so many
nations the world over, that the pristine age was one of un-
alloyed prosperity, peace and happiness, an Age of Gold, a
Saturnian Reign. Their legends asseverated that at that time
" the killing of a man was unknown, neither had there been
instances of their dying before they had attained to that age
which causes the hair to become white, the eyes dim, and the
teeth to be worn away."
This happy time was brought to a close by the advent of
1 " The Indians call the American continent an island; believing it to
be entirely surrounded by water." Heckewelder, Hist. Indian Nations,
p. 250.
2 Ibid, p. 308.
136 THE LENAPE AND THEIR LEGENDS.
certain evil beings who taught men how to kill each other by
sorcery/
Their kinsmen, the Mohegans, varied this cosmogonical
tradition, though retaining some of its main features. They
taught that in the beginning there was nought but water and
sky. At length from the sky a woman descended, our com-
mon mother. As she approached the boundless ocean, a small
point of land rose above the watery surface, and supplied her
with firm footing. She was pregnant by some mysterious
power, and she brought forth on this island animal triplets —
a bear, a deer and a wolf. From these all men and animals
are descended. The island grew to a main land, and the
mother of all, her mission accomplished, returned to her
home in the sky.'^
This creation-myth, obtained from the Indians around New
York harbor in the first generation after the advent of the
whites, has every mark of a genuine native production, and
coincides closely with that generally believed by the early
Algonkins.
It is followed by a migration myth, which ran to the effect
that their early forefathers came out of the northwest, for-
saking a tide-water country, and crossing over a great watery
tract, called ukhkok-pek, ''snake water, or water where snakes
are abundant," {dkhgook, snake, and pek, standing water,
1 Heckewelder, MSS. in the Library of the American Philosophical
Society. It is one of the points in favor of the authenticity of the Walam
Olum that this halcyon epoch is mentioned in its lines, though no reference
to it is contained in printed books relating to the Lenape legends.
2 Van der Donck, Description of the New Netherlands, Coll. N. Y.
Hist. Sac, Ser, II, Vol. I, pp. 217-18.
NATIONAL TRADITIONS. 137
probably from 7i'pey, water, akek, place or country). They
crossed many streams, but none in which the water ebbed
and flowed, until they reached the Hudson. ''Then they
said, one to another, ' This is like the Muhheakunnuck
(tidal ocean) of our nativity.' Therefore they agreed to
kindle a fire there and hang a kettle, whereof they and their
children after them might dip out their daily refreshment."
Hence came their name, the Tide-water People (see ante,
p. 20).
National Traditions.
Many early writers attest the passionate fondness of the
Delawares for their ancestral traditions and the memory of
their ancient heroes. The missionary, David Brainerd, men-
tions this as one of the leading difficulties in the way of
"evangelizing the Indians." "They are likewise much
attached," he writes, " to the traditions and fabulous notions
of their fathers, which they firmly believe, and thence look
upon their ancestors to have been the best of men."'^
To the same effect, Loskiel informs us that the Delawares
" love to relate what great warriors their ancestors had been,
and how many heroic deeds they had performed. It is a
pleasure to them to rehearse their genealogies. They are so
skilled at it that they can repeat the chief and collateral
lines with the utmost readiness. At the same time, they
characterize their ancestors, by describing this one as a
wise or skillful man, as a great chieftain, a renowned warrior,
a rich man, and the like. This they teach to their children,
^ Life and Journal of the Rev. David Brainerd, pp. 397, 425 (Edin-
burgh, 1826).
J
138 THE LENAPE AND THEIR LEGENDS.
and embody it i?i pictures, so as to make it more readily
remembered.''''^
The earliest writer who gives us any detailed description
of what these traditions were, is the Rev. Charles Beatty,
who visited the Delaware settlements in Ohio in 1767. On
his way there, he met a white man, Benjamin Sutton, who
for years had been a captive among the natives. He related
to Beatty the following tradition, which he had heard recited
by some old men among the Delawares: —
" That of old time their people were divided by a river,
nine parts of ten passing over the river, and one part remain-
ing behind; that they knew not, for certainty, how they
came to this continent ; but account thus for their first coming
into these parts where they are now settled ; that a king of
their nation, where they formerly lived, far to the west,
left his kingdom to his two sons ; that the one son making
war upon the other, the latter thereupon determined to depart
and seek some new habitation ; that accordingly he sat out,
accompanied by a number of his people, and that, after
wandering to and fro for the space of forty years, they at
length came to Delaware river, where they settled 370 years
ago. The way, he says, they keep an account of this is by
putting on a black bead of wampum every year on a belt
they keep for that purpose."^
^ So we may understand Loskiel to mean when he says, " Das bringen
sie ihren Kindern ebenfalls bey, und kleiden es in Bilder ein, um es noch
eindriicklicher zu machen." Geschichte der Mission, etc., s. 32. I think
Zeisberger, who was Loskiel's authority, meant Bilder in its literal, not
rhetorical, sense.
2 Qiarles Beatty, Journal of a Two Months'' Tour; with a View of
NATIONAL TRADITIONS. 139
From another source Mr. Beatty obtained the traditions
of the Nanticokes, which is apparently a version of that of
their relatives, the Delawares. It ran to this effect : At some
remote age, while on their way to their present homes, "They
came to a great water. One of the Indians that went before
them tried the depth of it by a long pole or reed, which he
had in his hand, and found it too deep for them to wade.
Upon their being non-plussed, and not knowing how to get
over it, their God made a bridge over the water in one night,
and the next morning, after they were all over, God took
away the bridge."^
A curious addition to this story is mentioned by Loskiel.^
The number of the mythical ancestors of their race who thus
were left on the shore of the great water was seven. This at
once recalls the seven c2ives{Chicomoztoc) or primitive stirpes
of the Mexican tribes, the seven clans {vuk amag) of the
Cakchiquels, the seven ancestors of the Qquechuas, etc., and
strongly intimates that there must be some common natural
occurrence to give rise to this wide-spread legend.^
Some peculiar sacredness must have attached to this number
among the Delawares also, as we are informed that the period
Promoting Religion among the Frontier Inhabitants of Pennsylvania,
and of Introducing Christianity among the Indians to the Westward of
the Alleghgeny Mountains, p. 27 (London, 1768).
1 Ibid, p. 91.
2 Geschichte der Mission, etc., p. 31.
3 The Mohegans seem also to have at one time had a sevenfold
division. At least a writer speaks of the "seven tribes" into which
those in Connecticut were divided. Mass. Hist. Soc. Colls., Vol. IX
(i ser.), p. 90.
140 THE LENAPE AND THEIR LEGENDS.
of isolation of their women at the catamenial period was seven
days/
The lunar month of 28 days, if divided and assigned equally
to each of the four cardinal points, would give a week of seven
days to each. Something of this kind seems to have been
done by another Algonkin tribe, the Ottawas, who declared
that the winds are caused (alternately ?) by seven genii or
gods who dwelt in the air.'^
The seven day period is also a natural, physical one, whose
influence is felt widely by vertebrate and invertebrate animals,
as Darwin has pointed out,^ and hence its appearance among
these people, who lived entirely subject to the operation of
their physical surroundings, is not so surprising.
The most complete account of the Delaware tradition is
that preserved by Heckewelder. In his pages it appears, not
as a reminiscence of tribal history, but as the tradition of the
whole eastern Algonkin race, and it claims for the three Dela-
ware tribes an antiquity of organization surpassing that of any
of their neighbors.
It holds such an important place that I quote all the essen-
tial passages : —
" The Lenni Lenape (according to the traditions handed
down to them by their ancestors) resided many hundred years
ago in a very distant country in the western part of the
American continent. For some reason, which I do not find
accounted for, they determined on migrating to the eastward,
and accordingly set out together in a body. After a very
1 Charles Beatty, Journal, etc., p. 84.
2 Relation des Jestiites, 1648, p. 77.
* The Descejit of Man, p. 165, note.
NATIONAL TRADITIONS. 141
long journey, and many nights' encampments by the way,
they at length arrived on the Namcesi Sipu, where they fell
in with the Mengwe, who had likewise emigrated from a dis-
tant country, and had struck upon this river somewhat higher
up. Their object was the same with that of the Delawares ;
they were proceeding on to the eastward, until they should
find a country that pleased therti. The spies which the Lenape
had sent forward for the purpose of reconnoitring, had long
before their arrival discovered that the country east of the
Mississippi was inhabited by a very powerful nation, who had
many large towns built on the great rivers flowing through
their land. Those people (as I was told) called themselves
Talligeu or Talligewi. Colonel John Gibson, however, a
gentleman who has a thorough knowledge of the Indians,'
and speaks several of their languages, is of opinion that they
were not called Talligewi, but Alligewi, * * *
" Many wonderful things are told of this famous people.
They are said to have been remarkably tall, and stout, and
there is a tradition that there were giants among them, people
of a much larger size than the tallest of the Lenape. It is
related that they had built to themselves regular fortifications
or entrenchments, from whence they would sally out, but
were generally repulsed. * * *
" When the Lenape arrived on the banks of the Mississippi,
they sent a message to the Alligewi to request permission to
settle themselves in their neighbourhood. This was refused
them, but they obtained leave to pass through the country
and seek a settlement farther to the eastward. They accord-
ingly began to cross the Namsesi Sipu, when the Alligewi,
seeing that their numbers were so very great, and in fact they
142 THE LENAPE AND THEIR LEGENDS.
consisted of many thousands, made a furious attack on those
who had crossed, threatening them all with destruction, if
they dared to persist in coming over to their side of the
river. * * *
" Having united their forces, the Lenape and Mengwe de-
clared war against the Alligewi, and great battles were fought,
in which many warriors fell on both sides. The enemy forti-
fied their large towns and erected fortifications, especially on
large rivers and near lakes, where they were successively at-
tacked and sometimes stormed by the allies. An engagement
took place in which hundreds fell, who were afterwards buried
in holes or laid together in heaps and covered over with earth.
No quarter was given, so that the Alligewi, at last, finding
that their destruction was inevitable if they persisted in their
obstinacy, abandoned the country to the conquerors, and
fled down the Mississippi river, from whence they never re-
turned. * * *
"In the end the conquerors divided the country between
themselves ; the Mengwe made choice of the lands in the
vicinity of the great lakes and on their tributary streams, and
the Lenape took possession of the country to the south. For
a long period of time — some say many hundred years — the
two nations resided peaceably in this country, and increased
very fast; some of their most enterprising huntsmen and
warriors crossed the great swamps, and falling on streams
running to the eastward, followed them down to the great
Bay river, thence into the Bay itself, which we call Chesa-
peak. As they pursued their travels, partly by land and
partly by water, sometimes near and at other times on the
great Salt-water Lake, as they call the sea, they discovered
NATIONAL TRADITIONS. 143
the great river, which we call the Delaware; and thence
exploring still eastward, the Scheyichbi country, now named
New Jersey, they arrived at another great stream, that which
we call the Hudson or North river. * * *
"At last they settled on the four great rivers (which we
call Delaware, Hudson, Susquehannah, Potomack), making
the Delaware, to which they gave the name of 'Lejiape-
wihittuck^ (the river or stream of the Lenape), the centre
of their possessions.
"They say, however, that the whole of their nation did
not reach this country ; that many remained behind, in order
to aid and assist that great body of their people which had
not crossed the Namsesi Sipu, but had retreated into the
interior of the country on the other side. * * *
"Their nation finally became divided into three separate
bodies ; the larger body, which they suppose to have been
one-half the whole, was settled on the Atlantic, and the other
half was again divided into two parts, one of which, the
strongest, as they suppose, remained beyond the Mississippi,
and the remainder where they left them, on this side of that
river.
"Those of the Delawares who fixed their abodes on the
shores of the Atlantic divided themselves into three tribes.
Two of them, distinguished by the names of the Turtle and
the Turkey, the former calling themselves Undmi, and the
other Unaldchtgo, chose those grounds to settle on which lay
nearest to the sea, between the coast and the high mountains.
As they multiplied, their settlements extended from the
Mohicanittuck (river of the Mohicans, which we call the
North or Hudson river) to the Potomack. * * *
144 THE LENAPE AND THEIR LEGENDS.
''The third tribe, the Wolf, commonly called 'Ccv^ Minsi,
which we have corrupted into Monseys, had chosen to live
back of the other two. * * * They extended their set-
tlements from the Minisink, a place named after them, where
they had their council seat and fire, quite up to the Hudson,
on the east ; and to the west or southward far beyond the
Susquehannah.
" From the above three tribes, the Unami, Unalachtgo and
the Minsi, had, in the course of time, sprung many others,
* * * the MahicaiiJii, or Mohicans, who spread themselves
over all that country which now composes the Eastern States,
* * * and the Nanticokes, who proceeded far to the south,
in Maryland and Virginia."
On their conquests during the period of their western mi-
grations, the Delawares based a claim for hunting grounds in
the Ohio valley. It is stated that when they had decided to
remove to the valley of the Muskingum, their chief, Neta-
watwes, presented this claim to the Hurons and Miamis, and
had it allowed.^ They also claimed lands on White River,
Indiana, and their settlement in that region at the close of
the last century was regarded as a return to their ancient
seats.
Nevertheless, in the earliest historic times, when the whites
first came in contact with the Lenape tribes, none of them
dwelt west of the mountains, nor, apparently, had they any
towns in the valley of the west branch of the Susquehanna
or of its main stream.
Although the above mentioned facts point to a migration
in prehistoric times from the West toward the East, there are
1 Heckewelder, Trans. Amer. Philos. Soc, Vol. IH, p. 388.
NATIONAL TRADITIONS. 145
indications of a yet older movement from the northeast west-
ward and southward to the upper Mississippi valley. A legend
common to the western Algonkin tribes, the Kikapoos, Sacs,
Foxes, Ottawas and Pottawatomies, located their original
home north of the St. Lawrence river, near or below where
Montreal now stands. In that distant land their ancestors
were created by the Great Spirit, and they dwelt there, " all
of one nation." Only when they removed or were driven
west did they separate into tribes speaking different dialects.^
The Shawnees, who at various times were in close relation
with the Delawares, also possessed a vague migration myth,
according to which, at some indefinitely remote past, they
had arrived at the main land after crossing a wide water.
Their ancestors succeeded in this by their great control of
magic arts, their occult power enabling them to walk over the
water as if it had been land. Until within the present century
this legend was repeated annually, and a yearly sacrifice offered
up in memory of their safe arrival.^ It is evidently a version
of that which appears in the third part of the Walam Olum.
1 This legend was told by the Sac Chief Masco, to Major Marston, about
1819. See J. Morse, Report on Indian Affairs, p. 138.
2 This myth was obtained in 18 12, from the Shawnees in Missouri
(Schoolcraft, Indian Tribes, Vol. IV, p. 254), and independently in 1819,
from those in Ohio (Mr. John Johnston, in Traits, of the Amer. Aniiq.
Soc, Vol. I, p. 273). Those of the tribe who now live on the Quapaw
Reservation, Indian Territory, repeat every year a long, probably mythical
and historical, chant, the words of which I have tried, in vain, to obtain.
They say that to repeat it to a white man would bring disasters on their
nation. I mention it as a piece of aboriginal composition most desirable
to secure.
146 THE LENAPE AND THEIR LEGENDS.
One of the curious legends of the Lenape was that of the
Great Naked or Hairless Bear. It is told by the Rev. John
Heckewelder, in a letter to Dr. B. S. Barton,^ The missionary
had heard it both among the Delawares and the Mohicans.
By the former, it was spoken of as amangachktidtmachque,
and in the dialect of the latter, ahamagachktidt mechqua^
The story told of it was that it was immense in size and
the most ferocious of animals. Its skin was bare, except a
tuft of white hair on its back. It attacked and ate the natives,
and the only means of escape from it was to take to the water.
Its sense of smell was remarkably keen, but its sight was
defective. As its heart was very small, it could not be easily
killed. The surest plan was to break its back-bone ; but so
dangerous was an encounter with it, that those hunters who
went in pursuit of it bade their families and friends farewell,
as if they never expected to return.
Fortunately, there were few of these beasts. The last one
known was to the east, somewhere beyond the left bank of the
Mahicanni Sipu (the Hudson river). When its presence was
learned a number of bold hunters went there, and mounted a
rock with precipitous sides. They then made a noise, and
attracted the bear's attention, who rushed to the attack with
great fury. As he could not climb the rock, he tore at it
1 Published in the Transactions of the American Philosophical Society,
1st ser., Vol. IV, pp. 260, sqq.
2 From amangi, great or big (in composition amangach), with the
accessory notion of terrible, or frightful ; Cree, amansis, to frighten ; tiat,
an abbreviated form of tmva, naked, whence the name Taiuaiaivas, or
Twightees, applied to the Miami Indians in the old records. (See Min-
utes of the Provincial Council of Penna., Vol. VIII, p. 41 8. J
NATIONAL TRADITIONS. 147
with his teeth, while the hunters above shot him with arrows
and threw upon him great stones, and thus killed him.
Though this was the last of the species, the Indian mothers
still used his name to frighten their children into obedience,
threatening them with the words, " The Naked Bear will eat
you."
CHAPTER VII.
The Walam Olum : Its Origin, Authenticity and
Contents.
Biographical Sketch of Rafinesque. — Value of his Writings. — His Account of the
Walam Olum. — Was it a Forgery? — Rafinesque's Character. — The Text pro-
nounced Genuine by Native Delawares. — Conclusion Reached.
Phonetic System of the Walam Olum — Metrical Form. — Pictographic System. —
Derivation and Precise Meaning of Walam Olum. — The MS. of the Walam
Olum. — General Synopsis of the Walum Olum. — Synopsis of its Parts.
Rafinesque and his Writings.
Constantine Samuel Rafinesque-Schmaltz, to whom we owe
the preservation and first translation of the Walam Olum,
was born in Galata, a suburb of Constantinople, Oct. 22d,
1783, and died in Philadelphia, of cancer of the stomach,
Sept. i8th, 1840.
His first visit to this country was in 1802. He remained
until 1804, when he went to Sicily, where he commenced
business. As the French were unpopular there, he added
"Schmaltz" to his name, for "prudent considerations,"
that being the surname of his mother's family.
In 1 815 he returned to America, but had the misfortune to
be shipwrecked on the coast, losing his manuscripts and much
of his property. On his arrival, he supported himself by
teaching, occupying his leisure time in scientific pursuits and
travel. In 1819 he was appointed "Professor of Historical
and Natural Sciences," in Transylvania University, Ken-
tucky.
148
RAFINESQUE AND HIS WRITINGS. 149
This position he was obliged to resign, for technical reasons,
in 1826, when he returned to Philadelphia, which city he
made his home during the rest of his life.
From his early youth he was an indefatigable student, col-
lector and writer in various branches of knowledge, especially
in natural history. On the title-page of the last work that
he published, "The Good Book and Amenities of Nature "
(Philadelphia, 1840), he claims to be the author of " 220 books,
pamphlets, essays and tracts." Including his contributions
to periodicals, there is no reason to doubt the correctness of
this estimate. They began when he was nineteen, and were
composed in English, French, Italian and Latin, all of which
he wrote with facility.
His earlier essays were principally on botanical subjects;
later, he included zoology and conchology ; and during the
last fifteen years of his life the history and antiquities of
America appear to have occupied his most earnest attention.
The value of his writings in these various branches has been
canvassed by several eminent critics in their respective lines.
First in point of time was Prof. Asa Gray, who in the year
following Rafinesque's death published in the "American
Journal of Science and Arts," Vol. XI, an analysis of his
botanical writings. He awards him considerable credit for
his earlier investigations, but much less for his later ones. To
quote Dr. Gray's words: "A gradual deterioration will be
observed in Rafinesque's botanical writings from 1819 to 1830,
when the passion for establishing new genera and species ap-
pears to have become a complete monomania. ' ' ^ But modern
believers in the doctrine of the evolution of plant forms and
1 American Journal of Science, Vol. XL, p. 237.
150 THE LENAPE AND THEIR LEGENDS,
the development of botanical species will incline to think
that there was a method in this madness, when they read the
passage from Rafinesque's writings, about 1836, which Dr.
Gray quotes as conclusively proving that, in things botanical,
Rafinesque had lost his wits. It is this : " But it is needless
to dispute about new genera, species and varieties. Every
variety is a deviation, which becomes a species as soon as it
is permanent by reproduction. Deviations in essential organs
may thus gradually become new genera." This is really an
anticipation of Darwinianism in botany.
The next year, in the same journal, appeared a " Notice of
the Zoological Writings of the late C. S. Rafinesque," by
Prof. S. S. Haldeman. It is, on the whole, depreciatory, and
convicts Rafinesque of errors of observation as well as of in-
ference; at the same time, not denying his enthusiasm and
his occasional quickness to appreciate zoological facts.
In 1864 the conchological writings of Rafinesque were
collected and published, in Philadelphia, by A. G. Binney
and Geo. W. Tryon, Jr., without comments. One of
the editors informs me that they have positive merit,
although the author was too credulous and too desirous
of novelties.
The antiquarian productions of Rafinesque, which interest
us most in this connection, were reviewed with caustic severity
by Dr. S. F. Haven, ^ especially the "Ancient Annals of
Kentucky," which was printed as an introduction to Mar-
shall's History of that State, in 1824. It is, indeed, an
absurd production, a reconstruction of alleged history on the,
flimsiest foundations ; but, alas ! not a whit more absurd than
1 Samuel F. Haven, Archaology of the United States, p. 40.
HISTORY OF THE WALAM OLUM. 151
the laborious card houses of many a subsequent antiquary of
renown.
His principal work in this branch appeared in Philadelphia
in 1836, entitled: "The American Nations; or, Outlines of
a National History ; of the Ancient and Modern Nations of
North and South America." It was printed for the author,
and is in two parts. Others were announced but never ap-
peared, nor did the maps and illustrations which the title
page promised. Its pages are filled with extravagant theories
and baseless analogies. In the first part he prints with notes
his translation of the Walam Olum, and his explanation of
its significance.
History of the Walam Olum.
Rafinesque's account of the origin of the Walam Olum
may be introduced by a passage in the last work he published,
" The Good Book." In that erratic volume he tells us that
he had long been collecting the signs and pictographs current
among the North American Indians, and adds : —
*' Of these I have now 60 used by the Southern or Floridian
Tribes of Louisiana to Florida, based upon their language of
Signs — 40 used by the Osages and Arkanzas, based on the
same — 74 used by the Lenapian (Delaware and akin) tribes
in their Wallamolum or Records — besides 30 simple signs
that can be traced out of the Neobagun or Delineation of the
Chipwas or Ninniwas, a branch of the last." ^
1 The Good Book ; or the Amenities of Nature. Printed for the
Eleutherium of Knowledge. Philadelphia, 1S40, pp. 77, 78. This
" Eleutherium," so far as I can learn, consisted of nobody but Monsieur
Rafinesque himself. Among his manifold projects was a " Divitial Sys-
152 THE LENAPE AND THEIR LEGENDS.
In these lines Rafinesque makes an important statement,
which has been amply verified by the investigations of Col.
Garrick Mallery, Dr. W. J. Hoffman and Capt. W. P. Clark,
within the last decade, and that is, that the Indian pictographic
system was based on their gesture speech.
So far as I remember, he was the first to perceive this sug-
gestive fact ; and he had announced it some time before 1840.
Already, in "The American Nations" (1836), he wrote,
" the Graphic Signs correspond to these Manual Signs." ^
Here he anticipates a leading result of the latest archseo-
logical research ; and I give his words the greater prominence,
because they seem to have been overlooked by all the recent
writers on Indian Gesture-speech and Sign-language.
The Neobagun, the Chipeway medicine song to which he
alludes, is likewise spoken of in "The American Nations,"
where he says : " The Ninniwas or Chipiwas * * have
such painted tales or annals, called Neobagun (male tool) by
the former. " ^ I suspect he derived his knowledge of this
from the Shawnee "Song for Medicine Hunting," called
" Nah-o-bah-e-gun-num," or. The Four Sticks, the words and
figures of which were appended by Dr. James to Tanner's
Narrative, published in 1830. ^
tern," by which all interested could soon become large capitalists. He
published a book on it (of course), which might be worth the attention of
a financial economist. The solid men of Philadelphia, however, like its
scholars, turned a deaf ear to the words of the eccentric foreigner.
^ The American Nations, etc., p. 78.
^ Ibid, p. X23.
3 Tanner's Narrative, p. 359.
DISCOVERY OF THE WALAM OLUM. 153
Discovery of the IValam Obcm.
As for the Lenape records, he gives this not very clear
account of his acquisition of them : —
"Having obtained, through the late Dr. Ward, of Indiana,
some of the original Wallam-Olura (painted record) of the
Linapi Tribe of Wapihani or White River, the translation
will be given of the songs annexed to each."^
On a later page he wrote : — ^
" Olum implies a record, a notched stick, an engraved
piece of wood or bark. It comes from ol, hollow or graved
record. * * * These actual olum were at first obtained
in 1820, as a reward for a medical cure, deemed a
curiosity; and were unexplicable. In 1822 were obtained
from another individual the songs annexed thereto in
the original language ; but no one could be found by me
able to translate them. I had therefore to learn the
language since, by the help of Zeisberger, Heckewelder
and a manuscript dictionary, on purpose to translate
them, which I only accomplished in 1833. The contents
were totally unknown to me in 1824, when I published
my 'Annals of Kentucky.' "
I have attempted to identify this ''Dr. Ward, of Indiana;"
but no such person is known in the early medical annals of
that State. There is, however, an old and well-known Ken-
tucky family of that name, who, about 1820, resided, and
still do reside, in the neighborhood of Cynthiana. One of
these, in 1824-25, was a friend of Rafinesque, invited him to
his house, and shared his archaeological tastes, as Rafinesque
^ American Natiotis, p. 122. ^ Ibid, p. 151.
K
154 THE LENAPE AND THEIR LEGENDS.
mentions in his autobiography.^ It was there, no doubt,
that he copied the signs and the original text of the Walam
Olum. My efforts to learn further about the originals from
living members of the family have been unsuccessful. From
a note in Rafinesque's handwriting, on the title page of his
MS. of 1833, it would appear that he had at least seen the
wooden tablets. This note reads : —
''This Mpt & the wooden original was (^/(:) procured in
1822 in Kentucky — but was inexplicable till a deep study of
the Linapi enabled me to translate them with explanations.
(Dr. Ward.)"
The name of Dr. Ward added in brackets is, I judge,
merely a note, and is not intended to imply that the sentence
is a quotation.
JVas it a Forgery?
The crucial question arises: Was the Walam Olum a
forgery by Rafinesque ?
It is necessary to ask and to answer this question, though
it seems, at first sight, an insult to the memory of the man
to do so. No one has ever felt it requisite to propound such
an inquiry about the pieces of the celebrated Mexican col-
lection of the Chevalier Boturini, who, as an antiquary, was
scarcely less visionary than Rafinesque.
But, unquestionably, an air of distrust and doubt shadowed
Rafinesque's scientific reputation during his life, and he was
not admitted on a favorable footing to the learned circles of
1 " My friend, Mr. Ward, took me to Cynthiana in a gig, where I sur-
veyed other ancient monuments." Rafinesque, A Life of Travels and
Researches,''' p. 74. (Phila., 1836.)
WAS IT A FORGERY ? 155
the city where he spent the last fifteen years of his life. His
articles were declined a hearing in its societies; and the
learned linguist, Mr. Peter Stephen Duponceau, whose spe-
cialty was the Delaware language, wholly and deliberately
ignored everything by the author of "The American
Nations."
Why was this ?
Rafinesque was poor, eccentric, negligent of his person,
full of impractical schemes and extravagant theories, and
manufactured and sold in a small way a secret nostrum which
he called " pulmel," for the cure of consumption. All these
were traits calculated to lower him in the respect of the citizens
of Philadelphia, and the consequence was, that although a
member of some scientific societies, he seems to have taken
no part in their proceedings, and was looked upon as an un-
desirable acquaintance, and as a sort of scientific outcast.
As early as 1819 Prof. Benjamin Silliman declined to publish
contributions from him in the ' 'American Journal of Science, ' ' ^
and returned him his MSS. Dr. Gray strongly intimates that
Rafinesque's assertions on scientific matters were at times in-
tentionally false, as when he said that he had seen Robin's
collection of Louisiana plants in France, whereas that botanist
never prepared dried specimens ; and the like.
I felt early in this investigation that Rafinesque's assertions
were, therefore, an insufficient warranty for the authenticity
of this document.
As I failed in my efforts to substantiate them by local re-
searches in Kentucky and Indiana, I saw that the evidence
must come from the text itself. Nor would it be sufficient to
1 Ajnerican Jotirnal of Science, Vol. XL, p. 237, note.
156 THE LENAPE AND THEIR LEGENDS.
prove that the words of the text were in the Lenape dialect.
With Zeisberger and Heckewelder at hand, both of whose
works had been years in print, it were easy to string together
Lenape words.
But what Rafinesque certainly had not the ability to do,
was to write a sentence in Lenape, to compose lines which an
educated native would recognize as in the syntax of his own
speech, though perhaps dialectically different.
This was the test that I determined to apply. I therefore
communicated my doubts to my friend, the distinguished
linguist, Mr. Horatio Hale, and asked him to state them to
the Rev. Albert Anthony, a well educated native Delaware,
equally conversant with his own tongue and with English.
Mr. Anthony considered the subject fully, and concluded
by expressing the positive opinion that the text as given was
a genuine oral composition of a Delaware Indian. In many
lines the etymology and syntax are correct ; in others there
are grammatical defects, which consist chiefly in the omission
of terminal inflections.
The suggestion he offered to explain these defects is ex-
tremely natural. The person who wrote down this oral
explanation of the signs, or, to speak more accurately, these
chants which the signs were intended to keep in memory, was
imperfectly acquainted with the native tongue, and did not
always catch terminal sounds. The speaker also may have
used here and there parts of that clipped language, or "white
man's Indian," which I have before referred to as serving for
the trading tongue between the two races.
This was also the opinion of the Moravian natives who ex-
amined the text. They all agreed that it impressed them as
WAS IT A FORGERY? 157
being of aboriginal origin, though the difference of the forms
of words left them often in the dark as to the meaning.
This very obscurity is in fact a proof that Rafinesque did
not manufacture it. Had he done so, he would have used the
'* Mission Delaware" words which he found in Zeisberger.
But the text has quite a number not in that dialect, nor in
any of the mission dictionaries.
Moreover, had he taken the words from such sources, he
would in his translation have given their correct meanings ;
but in many instances he is absurdly far from their sense.
Thus he writes: ''The word for angels, angelataiviwak, \%
not borrowed, but real Linapi, and is the same as the Greek
word angelos;''^ whereas it is a verbal with a future sense
from the very common Delaware verb angeln, to die. Many
such examples will be noted in the vocabulary on a later page.
In several cases the figures or symbols appear to me to
bear out the corrected translations which I have given of the
lines, and not that of Rafinesque. This, it will be observed,
is an evidence, not merely that he must have received this
text from other hands, but the figures also, and weighs
heavily in favor of the authentic character of both.
That it is a copy is also evident from some manifest mis-
takes in transcription, which Rafinesque preserves in his
printed version, and endeavored to translate, not perceiving
their erroneous form. Thus, in the fourth line of the first
chant, he wrote owak, translating it ''much air or clouds,"
when it is cleariy a mere transposition for woak, the Unami
form of the conjunction "and," as the sense requires. No
such blunder would appear if he had forged the document.
^ The American Nations, p. 151.
158 THE LENAPE AND THEIR LEGENDS.
It is true that a goodly share of the words in the earlier
chants occur in Zeisberger. Thus it seems, at first sight,
suspicious to find the three or four superlatives in III, 5, all
given under examples of the superlatives, in Zeisberger' s
Gramtnar, p. 105. It looks as if they had been bodily
transferred into the song. So I thought ; but afterwards I
found these same superlatives in Heckewelder, who added
specifically that ''the Delawares had formed them to address
or designate the Supreme being.'"
If we assume that this song is genuine, then Zeisberger
was undoubtedly familiar with some version of it ; had
learned it probably, and placed most of its words in his
vocabulary.
Some other collateral evidences of^ authenticity I have re-
ferred to on previous pages (pp. 67, 89, 136).
From these considerations, and from a study of the text,
the opinion I have formed of the Walam Olum is as fol-
lows : —
It is a genuine native production, which was repeated
orally to some one indifferently conversant with the Delaware
language, who wrote it down to the best of his ability. In
its present form it can, as a whole, lay no claim either to
antiquity, or to purity of linguistic form. Yet, as an authentic
modern version, slightly colored by European teachings, of
the ancient tribal traditions, it is well worth preservation,
and will repay more study in the future than is given it in
this volume. The narrator was probably one of the native
chiefs or priests, who had spent his life in the Ohio and
1 Correspondence between the Rev. John Heckewelder and Peter S.
Duponceau, Esq., p. 410.
METRICAL FORM. 159
Indiana towns of the Lenape, and who, though with some
knowledge of Christian instruction, preferred the pagan rites,
legends and myths of his ancestors. Probably certain lines
and passages were repeated in the archaic form in which
they had been handed down for generations.
Phonetic System.
The phonetic system adopted by the writer, whoever he
was, is not that of the Moravian brethren. They employed
the German alphabet, which does not obtain in the present
text. On this point Rafinesque says : " The orthography of
the Linapi names is reduced to the Spanish or French pro-
nunciation, except sh, as in English; u, as in French ; w, as
vcihow.'"'^ A comparison of the words with their equivalents
in Zeisberger's spelling shows that this is generally true.
It is obvious that the gutturals are few and soft, and that
the process of synthesis is carried further than in the Minsi
dialect. For this reason, from the introduction of peculiar
words, and from the loss of certain grammatical terminations,
the Minsi Delawares of to-day, to whom I have submitted it,
are of the opinion that it belongs to one of the southern
dialects of their nation ; perhaps to the Unalachtgo, as sug-
gested by Chief Gabriel Tobias, in his letter printed on a
preceding page (p. 88).
Metrical Form.
Even to an ear not acquainted with the language, the
chants of the Walam Glum are obviously in metrical arrange-
ment. The rhythm is syllabic and accentual, with frequent
^ The American Nations, p. 125.
160 THE LENAPE AND THEIR LEGENDS.
effort to select homophones (to which the correct form of
the words is occasionally sacrificed), and sometimes allitera-
tion. Iteration is also called in aid, and the metrical scheme
is varied in the different chants.
All these rhythmical devices appear in the native American
songs of many tribes, though I cannot point to any other
strictly aboriginal production in Algonkin, where a tendency
toward rhyme is as prominent as in the Walam Olum. It is
well to remember, however, that our material for comparison
is exceedingly scanty, and also that for nearly three-fourths
of a century before this song was obtained, the music-loving
Moravian missionaries had made the Delawares familiar with
numerous hymns in their own tongue, correctly framed and
rhymed.
Pictographic System.
The pictographic system which the Walam Olum presents
is clearly that of the Western Algonkins, most familiar to us
through examples from the Chipeways and Shawnees. It is
quite likely, indeed, that it was the work of a Shawnee, as
we know that they supplied such songs, with symbols, to the
Chipeways, and were intimately associated with the Dela-
wares.
At the time Rafinesque wrote, Tanner's Narrative had
been in print several years, and the numerous examples of
Algonkin pictography it contains were before him. Yet it
must be said that the pictographs of the Walam Olum have
less resemblance to these than to those published by the
Chipeway chief, George Copwa)^, in 1850, and by School-
craft, in his "History and Statistics of the Indian Tribes."
DERIVATION OF WALAM OLUM. 161
There is generally a distinct, obvious connection between
the symbol and the sense of the text, sufficient to recall the
latter to one who has made himself once thoroughly familiar
with it. I have not undertaken a study of the symbols ; but
have confined myself to a careful reproduction of them, and
the suggestion of their more obvious meanings, and their
correspondences with the pictographs furnished by later
writers. I shall leave it for others to determine to what
extent they should be accepted as a pure specimen of Algon-
kin pictographic writing.
Derivation of Walam Oliim.
The derivation of the name Walam Olum has been largely
anticipated on previous pages. I have shown that walam (in
modern Minsi, wdlumin) means ''painted," especially
"painted red.''^ This is a secondary meaning, as the root
Willi conveys the idea of something pleasant, in this con-
nection, pleasant to the eye, fine, pretty. (See ante p. 104.)
Olum was the name of the scores, marks, or figures in use
on the tally-sticks or record-boards. The native Delaware
missionary, Mr. Albert Anthony, says that the knowledge of
these ancient signs has been lost, but that the word olum is
still preserved by the Delaware boys in their games when
they keep the score by notches on a stick. These notches —
not the sticks — are called to this day olujn — an interesting
example of the preservation of an archaic form in the lan-
guage of children.
The name Wdldm Olum is therefore a highly appropriate
one for the record, and may be translated "Red Score."
162 THE LENAPE AND THEIR LEGENDS.
The MS. of the Walam Olum.
The MS. from which I have printed the Walam Olum is a
small quarto of forty unnumbered leaves, in the handwriting
of Rafinesque. It is in two parts with separate titles. The
first reads : —
Wallamolum.
First Part of the painted-engraved || traditions of the Linni linapi, &c. || containing || the
3 original traditional poems. || i. on the Creation and Ontogony, 24 verses. || 2. on the
Deluge, &c. 16 v. || 3. on the passage to America, 20 v. || Signs and Verses, 60 || with the
original glyphs or signs || for each verse of the poems or songs || translated word for
word II by C. S. Rafinesque || 1833.
The title of the second part is : —
Wallam-olum.
First and Second Parts of the || Painted and engraved traditions || of the Linni linapi.
IL Part.
Historical Chronicles or Annals || in two Chronicles.
I. From arrival in America to settlement in Ohio, &c. 4 chapters each of 16 verses,
each of 4 words, 64 signs.
2d. From Ohio to Atlantic States and back to Missouri, a mere succession of names
in 3 chapters of 20 verses — 60 signs.
Translated word for word by means of Zeisberger and Linapi Dictionary. With ex-
planations, &c.
By C. S. Rafinesque. 1833.
When Rafinesque died, his MSS. were scattered and passed
into various hands. Prof. Haldeman, in his notice above re-
ferred to (p. 150), stated that he and " Mr. Poulson of Phila-
delphia " had a large part of them.
This particular one, and also others descriptive of Rafin-
esque's arch^ological explorations in the southwest, his surveys
of the earthworks of Kentucky and the neighboring states,
and the draft of a work on " The Ancient Monuments of
North and South America," came into the possession of the
Hon. Brantz Mayer, of Baltimore, distinguished as an able
THE MS. OF THE WALUM OLUM. 163
public man and writer on American subjects, from whose
family I obtained them.
He loaned them all to Mr. E. G. Squier, who made exten-
sive use of Rafinesque's surveys, in the " Ancient Monuments
of the Mississippi Valley," giving due credit.
In June, 1848, Mr. Squier read before the New York His-
torical Society a paper entitled, " Historical and Mytho-
logical Traditions of the Algonquins ; with a translation of the
' Walum-Olum,' or Bark Record of the Linni-Lenape." This
was published in the "American Review," February, 1849,
and has been reprinted by Mr. W. W. Beach, in his " Indian
Miscellany" (Albany, 1877), and in the fifteenth edition of
Mr. S. G. Drake's "Aboriginal Races of North America."
This paper gave the symbols, original text and Rafinesque's
translation of the first two songs, and a free translation only,
of the remainder. The text was carelessly copied, whole
words being omitted, and no attempt was made to examine
the accuracy of the translation ; the symbols were also im-
perfect, several being reversed. Hence, as material for a
critical study of the document, Squier's essay is of little
value.
At the close of the second part of the MS. there are four
pages, closely written, with the title : —
" Fragment on the History of the Linapis since abt 1600 when the IVallmnolum
closes, translated from the Linapi by John Burns."
This was printed by Rafinesque and Squier, but as it has no
original text, as nothing is known of "John Burns," and as
the document itself, even if reasonably authentic, has no his-
toric value, I omit it.
164 THE LENAPE AND THEIR LEGENDS.
General Synopsis of the Walam Olujn.
The myths embodied in the earlier portion of the Walam
Olum are perfectly familiar to one acquainted with Algonkin
mythology. They are not of foreign origin, but are wholly
within the cycle of the most ancient legends of that stock.
Although they are not found elsewhere in the precise form
here presented, all the figures and all the leading incidents
recur in the native tales picked up by the Jesuit missionaries
in the seventeenth century, and by Schoolcraft, McKinney,
Tanner and others in later days.
In an earlier chapter I have collected the imperfect frag-
ments of these which we hear of among the Delawares, and
these are sufficient to show that they had substantially the
same mythology as their western relatives.
The cosmogony describes the formation of the world by
the Great Manito, and its subsequent despoliation by the
spirit of the waters, under the form of a serpent. The happy
days are depicted, when men lived without wars or sickness,
and food was at all times abundant. Evil beings, of mys-
terious power, introduced cold and war and sickness and
premature death. Then began strife and long wanderings.
However similar this general outline may be to European
and Oriental myths, it is neither derived originally from
them, nor was it acquired later by missionary influence.
This similarity is due wholly to the identity of psychological
action, the same ideas and fancies arising from similar im-
pressions in New as well as Old World tribes. No sound
ethnologist, no thorough student in comparative mythology,
would seek to maintain a genealogical relation of cultures on
GENERAL SYNOPSIS OF THE WALAM OLUM. 165
the strength of such identities. They are proofs of the
oneness of the human mind, and nothing more.
As to the historical portion of the document, it must be
judged by such corroborative evidence as we can glean from
other sources. I have quoted, in an earlier chapter, sufficient
testimony to show that the Lenape had traditions similar to
these, extending back for centuries, or at least believed by
their narrators to reach that far. What trust can be reposed
in them is for the archaeologist to judge.
Authentic history tells us nothing about the migrations of
the Lenape before we find them in the valley of the Delaware.
There is no positive evidence that they arrived there from
the west; still less concerning their earlier wanderings.
Were I to reconstruct their ancient history from the Walam
Olum, as I understand it, the result would read as follows: —
At some remote period their ancestors dwelt far to the
northeast, on tide-water, probably at Labrador (Compare
ante, p. 145). They journeyed south and west, till they
reached a broad water, full of islands and abounding in fish,
perhaps the St. Lawrence about the Thousand Isles. They
crossed and dwelt for some generations in the pine and
hemlock regions of New York, fighting more or less with the
Snake people, and the Talega, agricultural nations, living in
stationary villages to the southeast of them, in the area of
Ohio and Indiana. They drove out the former, but the latter
remained on the upper Ohio and its branches. The Lenape,
now settled on the streams in Indiana, wished to remove to
the East to join the Mohegans and other of their kin who had
moved there directly from northern New York. They, there-
fore, united with the Hurons (Talamatans) to drive out the
166 THE LENAPE AND THEIR LEGENDS.
Talega (Tsalaki, Cherokees) from the upper Ohio. This they
only succeeded in accomplishing finally in the historic period
(see ante p. 17). But they did clear the road and reached
the Delaware valley, though neither forgetting nor giving up
their claims to their western territories (see ante p. 144).
In the sixteenth century the Iroquois tribes seized and oc-
cupied the whole of the Susquehanna valley, thus cutting off
the eastern from the western Algonkins, and ended by driving
many of the Lenape from the west to the east bank of the
Delaware (ante p. ;^8).
Synopsis of the separate parts.
I.
The formation of the universe by the Great Manito is de-
scribed. In the primal fog and watery waste he formed land
and sky, and the heavens cleared. He then created men and
animals. These lived in peace and joy until a certain evil
manito came, and sowed discord and misery.
This canto is a version of the Delaware tradition mentioned
in the Heckewelder MSS. which I have given previously, p.
135. The notion of the earth rising from the prim.al waters
is strictly a part of the earliest Algonkin mythology, as I have
amply shown in previous discussions of the subject. See my
Myths of the New World, p. 213, and American Hero Myths,
Chap. II.
II.
The Evil Manito, who now appears under the guise of a
gigantic serpent, determines to destroy the human race, and
for that purpose brings upon them a flood of water. Many
perish, but a certain number escape to the turtle, that is, to
solid land, and are there protected by Nanabush (Manibozho
SYNOPSIS OF THE SEPARATE PARTS. 167
or Michabo). They pray to him for assistance, and he caused
the water to disappear, and the great serpent to depart.
This canto is a brief reference to the conflict between the
Algonkin hero god and the serpent of the waters, originally,
doubtless, a meteorological myth. It is an ancient and
authentic aboriginal legend, shared both by Iroquois and
Algonkins, under slightly different forms. In one aspect, it
is the Flood or Deluge Myth. For the general form of this
myth, see my Myths of the New World, pp. 119, 143, 182,
and American Hero Myths, p. 50, and authorities there
quoted; also, E. G. Squier, " Manabozho and the Great
Serpent; an Algonquin Tradition," in \hQ American Review,
Vol. II, Oct., 1848.
III.
The waters having disappeared, the home of the tribe is
described as in a cold northern clime. This they concluded
to leave in search of warmer lands. Having divided their
people into a warrior and a peaceful class, they journeyed
southward, toward what is called the "Snake land." They
approached this land in winter, over a frozen river. Their
number was large, but all had not joined in the expedition
with equal willingness, their members at the west preferring
their ancient seats in the north to the uncertainty of southern
conquests. They, however, finally united with the other
bands, and they all moved south to the land of spruce pines.
IV.
The first sixteen verses record the gradual conquest of
most of the Snake land. It seems to have required the suc-
cessive efforts of six or seven head chiefs, one after another,
168 THE LENAPE AND THEIR LEGENDS.
to bring this about, probably but a small portion at a time
yielding to the attacks of these enemies. Its position is
described as being to the southwest, and in the interior of
the country. Here they first learned to cultivate maize.
The remainder of the canto is taken up with a long list of
chiefs, and with the removal of the tribe, in separate bands
and at different times, to the east. In this journey from the
Snake land to the east, they encountered and had long wars
with the Talega. These lived in strong towns, but by the
aid of the Hurons (Talamatans), they overcame them and
drove them to the south.
V.
Having conquered the Talegas, the Lenape possessed their
land and that of the Snake people, and for a certain time
enjoyed peace and abundance. Then occurred a division of
their people, some, as Nanticokes and Shawnees, going to
the south, others to the west, and later, the majority toward
the east, arriving finally at the Salt sea, the Atlantic ocean.
Thence a portion turned north and east, and encountered
the Iroquois. Still later, the three sub-tribes of the Lenape
settled themselves definitely along the Delaware river, and
received the geographical names by which they were known,
as Minsi, Unami and Unalachtgo (see ante, p. 36). They
were often at war with the Iroquois, generally successfully.
Rumors of the whites had reached them, and finally these
strangers approached the river, both from the north (New
York bay) and the south. Here the song closes.
THE WALAM OLUM,
RED SCORE,
LENAPE
170 THE LENAPE AND THEIR LEGENDS.
I.
I. Sayewi talli wemiguma wok-
getaki,
2. Hackung kwelik owanaku
wak yutali Kitanitowit-
essop.
3. Sayewis hallemiwis nolemiwi
elemamik Kitanitowit-es-
sop.
4. Sohalawak kwelik hakik
owak^ awasagamak.
O^:
5. Sohalawak gishuk nipahum
alankwak.
6. Wemi-sohalawakyulikyuch-
aan.
7. Wich-owagan kshakan mo-
shakwaf kwelik kshipe-
helep.
8, Opeleken mani-menak del-
sin-epit.
1 Read, woak.
2 Var. moshakguat.
THE WALAM OLUM. 171
I.
I. At first, in that place, at all times, above the earth,
2. On the earth, [was] an extended fog, and there the
great Manito was.
3. At first, forever, lost in space, everywhere, the great
Manito was.
4. He made the extended land and the sky.
5. He made the sun, the moon, the stars.
6. He made them all to move evenly.
7. Then the wind blew violently, and it cleared, and the
water flowed off far and strong.
8. And groups of islands grew newly, and there re-
mained.
172 THE LENAPE AND THEIR LEGENDS.
9. Lappinup Kitanitowit manito
manitoak.
10. Owiniwak angelatawiwak
chichankwak wemiwak.
II. Wtenk manito jinwis lenno-
wak mukom.
12. Milap netami gaho owini
gaho.
:^
^
•SIS
^
13. Namesik milap, tulpewik mi-
lap, awesik milap, cholen-
sak milap.
14. Makimani shak sohalawak
makowini nakowak aman-
gamek.
THE WALAM OLUM. 173
9. Anew spoke the great Manito, a manito to manitos,
10. To beings, mortals, souls and all,
II. And ever after he was a manito to men, and their
grandfather.
12. He gave the first mother, the mother of beings.
13. He gave the fish, he gave the turtles, he gave the
beasts, he gave the birds.
14. But an evil Manito made evil beings only, monsters,
174 THE LENAPE AND THEIR LEGENDS.
15. Sohalawak uchewak, sohala-
wak pungusak.
r\
o o
A
o o
— ^37 —
16. Nitisak wemi owini w'delsi-
newuap.
'h^^
17. Kiwis, wunand wishimani-
^ toak essopak.
18. Nijini netami lennowak, ni-
goha netami okwewi, nan-
tinewak.
19. Gattamin netami mitzi nijini
nantine.
u
20. Wemi wingi-namenep, wemi
ksin-elendamep, wemi wul-
latemanuwi.
' 1~^ — 21. Shukandeli-kimimekenikink
J^^^^'^^'-^ wakon powako init'ako.
THE WALAM OLUM. 175
15. He made the flies, he made the gnats.
16. All beings were then friendly.
17. Truly the manitos were active and kindly
18. To those very first men, and to those first mothers;
fetched them wives,
19. And fetched them food, when first they desired it.
20. All had cheerful knowledge, all had leisure, all
thought in gladness.
21. But very secretly an evil being, a mighty magician,
came on earth,
176
THE LENAPE AND THEIR LEGENDS.
O* A
J^^^A_<
W
0^=°""^—
^.
22. Mattalogas pallalogas mak-
taton owagan payat-chik
yutali.
23. Maktapan payat, wihillan
payat, mboagan payat.
24. Won wemi wiwunch kamik
atak kitahikan netamaki
epit.
II.
I. Wulamo maskanako anup
lennowak makowini esso-
pak.
2. Maskanako shingalusit nijini
essopak shawelendamep
eken shingalan.
3. Nishawi palHton, nishawi
machiton, nishawi matta
lungundowin.
4. Mattapewi wiki nihanlowit
mekwazoan.
THE WALAM OLUM. 177
22. And with him brought badness, quarreling, unhap-
piness,
23. Brought bad weather, brought sickness, brought
death.
24. All this took place of old on the earth, beyond the
great tide-water, at the first.
11.
I . Long ago there was a mighty snake and beings evil
to men.
2. This mighty snake hated those who were there (and)
greatly disquieted those whom he hated.
3. They both did harm, they both injured each other,
both were not in peace.
4. Driven from their homes they fought with this mur-
derer.
178 THE LENAPE AND THEIR LEGENDS.
5. Maskanako gishi penauwe-
lendamep lennowak owini
palliton.
6. Nakowa petonep, amangam
petonep, akopehella peto-
nep.
7. Pehella pehella, pohoka po-
hoka, eshohok eshohok,
palliton palliton.
8, Tulapit menapit Nanaboush
maskaboush owinimokom
linowimokom.
9. Gishikin-pommixin tulagis-
hatten-lohxin.
10. Owini linowi wemoltin,
Pehella gahani pommixin,
Nahiwi tatalli tulapin.
THE WALAM OLUM. 179
5. The mighty snake firmly resolved to harm the men.
6. He brought three persons, he brought a monster, he
brought a rushine water.
7. Between the hills the water rushed and rushed, dash-
ing through and through, destroying much.
8. Nanabush, the Strong White One, grandfather of
beings, grandfather of men, was on the Turtle
Island.
9. There he was walking and creating, as he passed by
and created the turtle.
10. Beings and men all go forth, they walk in the floods
and shallow waters, down stream thither to the
Turtle Island.
180 THE LENAPE AND THEIR LEGENDS.
(:-vl SIltttn>, ^^* Amanganek makdopannek
-jI — _■■ i TT ""^^ alendvuwek metzipannek.
12. Manito-dasin mokol-wiche-
map,
Palpal payat payat wemiche-
map.
13. Nanaboush Nanaboush we-
mimokom,
Winimokom linnimokom tu-
lamokom.
14. Linapi-ma tulapi-ma tulape-
wi tapitawi.
15. Wishanem tulpewi pataman
tulpewi poniton wuliton.
16. Kshipehelen penkwihilen,
Kwamipokho sitwalikho,
Maskan wagan palliwi pal-
liwi.
n
QL
<vr^
^.^
X
III.
1. Pehella wtenk lennapewi tu-
lapewini psakwiken woli-
wikgun wittank talli.
2. Topan-akpinep, wineu-akpi-
nep, kshakan-akpinep, thu-
pin akpinep.
THE WALAM OLUM. 181
II. There were many monster fishes, which ate some of
them.
12. The Manito daughter, coming, helped with her canoe,
helped all, as they came and came.
13. [And also] Nanabush, Nanabush, the grandfather of
all, the grandfather of beings, the grandfather of
men, the grandfather of the turtle.
14. The men then were together on the turtle, like to
turtles.
15. Frightened on the turtle, they prayed on the turtle
that what was spoiled should be restored.
16. The water ran off, the earth dried, the lakes were at
rest, all was silent, and the mighty snake departed.
III.
1. After the rushing waters (had subsided) the Lenape
of the turtle were close together, in hollow houses,
living together there.
2. It freezes where they abode, it snows where they
abode, it storms where they abode, it is cold where
they abode.
182
THE LENAPE AND THEIR LEGENDS.
^=b
3. Lowankwamink wulaton
wtakan tihill kelik me-
shautang sili ewak.
4. Chintanes-sin powalessin
peyachik wikhichik pok-
wihil.
5. Eluwi-chitanesit eluwi takau-
wesit, elowi chiksit, elowi-
chik delsinewo.
6. Lowaniwi, wapaniwi, shawa-
niwi, wunkeniwi, elowichik
apakachik.
7. Lumowaki, lowanaki tulpe-
naki elowaki tulapiwi lina-
piwi.
8. Wemiako yagawan tendki
lakkawelendam nakopowa
wemi owenluen atam.
9. Akhokink wapaneu wemoltin
palliaal kitelendam apte-
lendam.
THE WALAM OLUM. 183
3. At this northern place they speak favorably of mild,
cool (lands), with many deer and buffaloes.
4. As they journeyed, some being strong, some rich,
they separated into house-builders and hunters;
5. The strongest, the most united, the purest, were the
hunters.
6. The hunters showed themselves at the north, at the
east, at the south, at the west.
7. In that ancient country, in that northern country, in
that turtle country, the best of the Lenape were
the Turtle men.
8. All the cabin fires of that land were disquieted, and
all said to their priest, " Let us go."
9. To the Snake land to the east they went forth, going
away, earnestly grieving.
184 THE LENAPE AND THEIR LEGENDS.
-^. 10. Pechimuin shakowen^ nungi-
YJ-fx"^ hillan lusasaki pikihil pok-
j — ^- ' wihil pkomenpti
II. Nihillapewin komelendam
M W C/ lowaniwi wemiten chihillen
\ \ 'V winiakpn
-^ 12. Namesuagipek pokhapock-
hapek guneunga wapla-
newa ouken waptumewi
ouken.
1 3. Amokolon nallahemen agun-
ouken pawasinep wapasi-
nep akomenep.'^
//-^J^l) (P- 14. Wihlamokkicholenluchundi,
Wematam akomen luchundi.
V >^^^-==ccII7t^ 15. Witehen wemiluen wemaken
ti^______P~ nihillen.
^ 16. Nguttichin lowaniwi,
Y yi^ ^_ Nguttichin wapaniwi,
p*^-^ \ — r' .^^^^ Agamunk topanpek
^ Wulliton epannek.
Arf ,5r5 I/' Wulelemil w'shakuppek,
[1::nJ_[ oJjT ^ Wemopannek hakhsinipek,
^""--^ — ^" Kitahikan pokhakhopek.
1 Var. showoken. 1 Var. itienakmep.
THE WALAM OLUM. 185
lO. Split asunder, weak, trembling, their land burned,
they went, torn and broken, to the Snake Island.
II. Those from the north being free, without care, went
forth from the land of snow, in different directions.
1 2. The fathers of the Bald Eagle and the White Wolf
remain along the sea, rich in fish and muscles.
13. Floating lip the streams in their canoes, our fathers
were rich, they were in the light, when they were
at those islands.
14. Head Beaver and Big Bird said,
" Let us go to Snake Island," they said.
15. All say they will go along to destroy all the land.
16. Those of the north agreed.
Those of the east agreed.
Over the water, the frozen sea.
They went to enjoy it.
17. On the wonderful, slippery water,
On the stone-hard water all went.
On the great Tidal Sea, the muscle-bearing sea.
M
186
THE LENAPE AND THEIR LEGENDS.
PNJI^I^^
1 8. Tellenchen kittapakki nillawi,
Wemoltin gutikuni nillawi,
Akomen wapanawaki nillawi,
Ponskan, ponskan, wemiwi
olini.
19. Lowanapi, wapanapi, shawa-
napi,
Lanewapi, tamakwapi, tume-
wapi,
Elowapi, powatapi, wilawapi,
Okwisapi, danisapi, allumapi.
X A
20. Wemipayat guneunga shina-
king,
Wunkenapi chanelendam
payaking,
Allowelendam kowiyey tul-
paking.
IV.
A
^
A.
I. Wulamo linapioken manup
shinaking.
2. Wapallanewa sittamaganat
yukepechi wemima,
3. Akhomenis michihaki wel-
laki kundokanup.
THE WALAM OLUM. 187
1 8. Ten thousand at night,
All in one night,
To the Snake Island, to the east, at night,
They walk and walk, all of them.
19. The men from the north, the east, the south,
The Eagle clan, the Beaver clan, the Wolf clan.
The best men, the rich m.en, the head men,
Those with wives, those with daughters, those with
dogs.
20. They all come, they tarry at the land of the spruce
pines;
Those from the west come with hesitation.
Esteeming highly their old home at the Turtle land.
IV.
I. Long ago the fathers of the Lenape were at the land
of spruce pines.
2, Hitherto the Bald Eagle band had been the pipe
bearer.
3. While they were searching for the Snake Island, that
great and fine land.
188 THE LENAPE AND THEIR LEGENDS.
4. Angomelchik elowichik el-
musichik menalting.
5. Wemilo kolawil sakima lis-
silma.
6. Akhopayat kihillalend akho-
pokho askiwaal.
7. Showihilla akhowemi gand-
haton mashkipokhing.
#
8, Wtenkolawil shinaking saki-
manep wapagokhos.
9. Wtenk nekama sakimanep
janotowi enolowin.
10. Wtenk nekama sakimanep
chilili shawaniluen.
THE WALUM OLUM. 189
4. They having died, the hunters, about to depart, met
together.
5. All say to Beautiful Head, "Be thou chief."
6. " Coming to the Snakes, slaughter at that Snake hill,
that they leave it,"
7. All of the Snake tribe were weak, and hid them-
selves in the Swampy Vales.
8. After Beautiful Head, White Owl was chief at Spruce
Pine land.
9. After him, Keeping-Guard was chief of that people.
10. After him. Snow Bird was chief; he spoke of the
south,
190 THE LENAPE AND THEIR LEGENDS.
II, Wokenapi nitaton wullaton
apakchikton.
r
12. Shawaniwaen chilili, wapani-
waen tamakwi.
13. Akolaki shawanaki, kitshi-
naki shabiyaki.
14. Wapanaki namesaki, pema-
paki sisilaki.
/
1 5 , Wtenk chilili sakimanep aya-
mek weminilluk.
16. Chikonapi akhonapi maka-
tapi assinapi.
17. Wtenk ayamek tellen saki-
mak machi tonanup shawa-
pama.
THE WALAM OLUM. 191
II. That our fathers should possess it by scattering
abroad.
12. Snow Bird went south, White Beaver went east.
13. The Snake land was at the south, the great Spruce
Pine land was toward the shore ;
14. To the east was the Fish land, toward the lakes was
the buffalo land.
15, After Snow Bird, the Seizer was chief, and all were
killed,
16. The robbers, the snakes, the evil men, the stone men.
17. After the Seizer there were ten chiefs, and there was
much warfare south and east.
192 THE LENAPE AND THEIR LEGENDS.^
x^^
1 8. Wtenk nellamawa sakimanep
langundowi akolaking.
19. Wtenk nekama sakimanep
tasukamend shakagapipi.
f
20. Wtenk nekama sakimanep
pemaholend wulitowin.
e^l^
21. Sagimawtenk matemik, sagi-
mawtenk pilsohalin.
22. Sagimawtenk gunokeni, sagi-
mawtenk mangipitak.
Jl
23. Sagimawtenk olumapi, lek-
sahowen sohalawak.
V
24, Sagimawtenk taguachi sha-
waniwaen minihaking.
25. Sakimawtenk huminiend mi-
nigeman sohalgol.
THE WALAM OLUM. 193
1 8. After them, the Peaceable was chief at Snake land.
19. After him, Not-Black was chief, who was a straight
man.
20. After him, Much-Loved was chief, a good man.
21. After him, No-Blood was chief, who walked in clean-
liness.
22. After him, Snow-Father was chief, he of the big
teeth.
23. After him, Tally-Maker was chief, who made records.
24. After him, Shiverer-with-Cold was chief, who went
south to the corn land.
25. After him, Corn-Breaker was chief, who brought
about the planting of corn.
194
THE LENAPE AND THEIR LEGENDS.
zzJz
26. Sakimawtenk alkosohit saki-
machik apendawi.
A i/C 27. Sawkimawtenk shiwapi, saki-
1 I n 1 1 ■ ' ■ '^ matenk penkwonwi.
3-
6
28. Attasokelaii attaminin wapa-
niwaen italissipek.
£&
29. Oligonunk sisilaking nalli-
metzin kolakwaming.
^ i
30. Wtenk penkwonwi wekwo-
chella, wtenk nekama chin-
ofalsuwi.
31. Wtenk nekama kwitikwond,
slangelendam attagatta,
32. Wundanuksin wapanickam^
allendyachick kimimikwi.
33. Gunehunga wetatamowi wa-
kaholend sakimalanop.
^ Var. zvapanahan.
THE WALAM OLUM. 195
26. After him, the Strong-Man was chief, who was use-
ful to the chieftains.
27. After him, the Salt-Man was chief; after him the
Little-One was chief
28. There was no rain, and no corn, so they moved fur-
ther seaward.
29. At the place of caves, in the buffalo land, they at
last had food, on a pleasant plain.
30. After the Little-One (came) the Fatigued ; after him,
the Stiff-One.
3 1 . After him, the Reprover ; disliking him, and unwil-
ling (to remain).
32. Being angry, some went off secretly, moving east.
33. The wise ones who remained made the Loving-One
chief
196 THE LENAPE AND THEIR LEGENDS.
34. Wisawana lappi wittank mi-
chi mini madawasim.
35. Weminitis tamenend sakima-
nep nekohatami.
36. Eluwiwulit matemenend we-
mi linapi nitis payat.
37. Wtenk wulitma maskansisil
y sakimanep w'tamaganat.
38. Machigokloos sakimanep,
wapkicholen sakimanep.
39. Wingenund sakimanep po-
watanep gentikalanep.
40, Lapawin sakimanep, wallama
sakimanep.
^■^x'
V •
1:^ V
41. Waptipatit sakimanep, lappi
mahuk lowashawa.
THE WALAM OLUM. 197
34. They settled again on the Yellow river, and had
much corn on stoneless soil.
35. All being friendly, the Affable was chief, the first of
that name.
36. He was very good, this Affable, and came as a friend
to all the Lenape.
37. After this good one, Strong-Buffalo was chief and
pipe-bearer.
38. Big-Owl was chief; White-Bird was chief.
39. The Willing-One was chief and priest ; he made fes-
tivals.
40. Rich-Again was chief; the Painted-One was chief
41. White-Fowl was chief; again there was war, north
and south.
198 THE LENAPE AND THEIR LEGENDS.
42. Wewoattan menatting tuma-
j4) okan sakimanep.
^Ji / /\
^■
43. NitatonepwemipalHtonmas-
kansini nihillanep.
(q)' J^ 44- Messissuwi sakimanep ako
— 4-:='Q — wini oallitoneo.
wini pallitonep.
45, Chitanwulit sakimanep lowa-
nuski pallitonep.
Q 46. Alokuwi sakimanep towakon
"D ><n pallitonep.
47. Opekasit sakimanep sakhe-
lendam pallitonepit.
/yy/7^ J)^-"-^ 4^' Wapagishik yuknohokluen
j nonon o_A makeluhuk wapaneken.
6
6'
[\ )Lzi2i- 49- Tsehepieken nemassipi^ no-
^vC landowak gunehunga.
^ Var. mixtisipi.
THE WALAM OLUM. 199
42. The Wolf-wise-in-Counsel was chief.
43. He knew how to make war on all ; he slew Strong-
Stone.
44, The Always-Ready-One was chief; he fought against
the Snakes.
45. The Strong-Good-One was chief; he fought against
the northerners.
46. The Lean-One was chief; he fought against the
Tawa people.
47. The Opossum-Like was chief; he fought in sadness,
48. And said, " They are many ; let us go together to
the east, to the sunrise."
49. They separated at Fish river ; the lazy ones remained
there.
200 THE LENAPE AND THEIR LEGENDS.
^^-zS
Dr-
1^
,x
50. Yagawanend sakimaiiep tal-
ligewi wapawullaton.
^ 51. Chitanitis sakimanep wapa-
^ — Y waki gotatamen.
52. Wapallendi pomisinep tale-
gawil allendhilla.
53. Mayoksuwi wemilowi palH-
ton palHton.
54. Talamatan nitilowan payat-
chik wemiten.
JA" ^-y- 55. Kinehepend sakimanep ta-
\ r'^v^ J maganat sipakgamen.
j^ j2l7~ 5 6- Wulatonwi makelima palli-
— '^^Sf^^ hilla taleg-awik.
57. Pimokhasuwi sakimanep
wsamimaskan talesrawik.
O r9 58. Tenchekentit sakimanep we-
' milat makelinik.
THE WALAM OLUM. 201
50. Cabin-Man was chief; the Talligewi possessed the
east.
5 1. Strong-Friend was chief; he desired the eastern land.
52. Some passed on east; the Talega ruler killed some
of them.
53. All say, in unison, "War, war."
54. The Talamatan, friends from the north, come, and
all go together.
55. The Sharp-One was chief; he was the pipe-bearer
beyond the river.
56. They rejoiced greatly that they should fight and slay
the Talega towns.
57. The Stirrer was chief; the Talega towns were too
strong.
58. The Fire-Builder was chief; they all gave to him
many towns.
202 THE LENAPE AND THEIR LEGENDS.
^_^ 59. Pagan chihilla sakimanep
^^ "^ shawanewak wemi talega.
T
/T
60. Hattan wulaton sakimanep,
wingelendam wemi lenno-
A 7) K 61. Shawanipekis gunehungind
}A^-^-'^}Ti lowanipekis talamatanitis.
62. Attabchinitis gishelendam
gunitakan sakimanep.
63. Linniwulamen sakimanep
pallitonep talamatan.
64. Shakagapewi sakimanep nun-
giwi talamatan.
V.
%£
T@3-
1. Wemilangundo wulamo talli
talegaking.
2. Tamaganend sakimanep wa-
palaneng.
/^M/V/7^ 3. Wapushuwi sakimanep kelit-
L ;/////// geman.
THE WALAM OLUM. 203
59. The Breaker-in-Pieces was chief; all the Talega go
south.
60. He-has-Pleasure was chief; all the people rejoice.
61. They stay south of the lakes ; the Talamatan friends
north of the lakes.
62. When Long-and-Mild was chief, those who were not
his friends conspired.
63. Truthful-Man was chief; the Talamatans made war.
64. Just-and-True was chief; the Talamatans trembled.
V.
1. All were peaceful, long ago, there at the Talega land.
2. The Pipe-Bearer was chief at the White river.
3. White-Lynx was chief; much corn was planted.
204 THE LENAPE AND THEIR LEGENDS.
^
66666A
4. Wulitshinik sakimanep mak-
dopannik.
5. Lekhihitin sakimanep wal-
lamolumin.
6. Kolachuisen sakimanep ma-
keliminp'.
n nn
7. Pematalli sakimanep make-
linik.
8. Pepomahenem sakimanep
makelaning.
9. Tankawon sakimanep make-
leyachik,
10. Nentegowi shawanowi sha-
wanaking.
II. Kichitamak sakimanep wa-
pahoning.
a
12. Onowutok awolagan wunke-
nahep.
w^.
13. Wunpakitonis wunshawono-
nis wunkiwikwotank.
THE WALAM OLUM. 205
4. Good-and-Strong was chief; the people were many.
5. The Recorder was chief; he painted the records,
6. Pretty-Blue-Bird was chief; there was much fruit.
7. Al ways-There was chief; the towns were many.
8. Paddler-up-Stream was chief; he was much on the
rivers.
9. Little-Cloud was chief; many departed,
10. The Nanticokes and the Shawn ees going to the
south.
II. Big-Beaver was chief, at the White Salt Lick.
12. The Seer, the praised one, went to the west.
13. He went to the west, to the southwest, to the western
villages.
206 THE LENAPE AND THEIR LEGENDS.
r 14. Pawanami sakimanep tale-
-^ ganah.
X X
n" 15. Lokwelend sakimanep mak-
/ \ nplliton
palliton
■s^^S^ \/ 16. Lappi towako lappi sinako
^ ^^ lappi lowako.
yS\>^
17. Mokolmokom sakimanep
Od-^ mokolakolin.
,^
c
18. Winelowich sakimanep lo-
wushkakiang.
19. Linkwekinuk sakimanep ta-
legachukang.
20. Wapalawikwan sakimanep
waptalegawing.
21. Amangaki amigaki wapaki-
sinep.
fTTh 1 1 / n 22, Mattakohaki mapawaki ma-
X wulitenol.
THE WALAM OLUM. 207
14. The Rich-Down-River-Man was chief, at Talega
river.
15. The Walker was chief; there was much war.
16. Again with the Tawa people, again with the Stone
people, again with the northern people.
17. Grandfather-of-Boats was chief; he went to lands in
boats.
18. Snow-Hunter was chief; he went to the north land.
19. Look-About was chief; he went to the Talega moun-
mountains.
20. East-Villager was chief; he was east of Talega.
21. A great land and a wide land was the east land,
22. A land without snakes, a rich land, a pleasant land.
208 THE LENAPE AND THEIR LEGENDS.
■^ 23. Gikenopalat sakimanep pe-
kochilowan.
j/ 24. Saskwihanang hanaholend
AWb ^-^'C^'' sakimanep.
A
_^^-
25. Gattawisi sakimanep wina-
kaking.
26. Wemi lowichik gishiksha-
wipek lappi kichipek.
/^
ftzi
27. Makhiawip sakimanep lapi-
haneng.
28. Wolomenap sakimanep mas-
kekitong.
29. Wapanand tumewand waplo-
waan.
30. Wulitpallat sakimanep pisk-
wilowan.
31. Mahongwi pungelika wemi
nungwi.
THE WALAM OLUM. 209
23. Great Fighter was chief, toward the north,
24. At the Straight river, River-Loving was chief.
25. Becoming-Fat was chief at Sassafras land.
26. All the hunters made wampum again at the great sea.
27. Red-Arrow was chief at the stream again.
28. The Painted-Man was chief at the Mighty Water.
29. The Easterners and the Wolves go northeast.
30. Good-Fighter was chief, and went to the north.
31. The Mengwe, the Lynxes, all trembled.
210 THE LENAPE AND THEIR LEGENDS.
32. Lappi tamenend sakimanepit
wemi langundit.
33. Wemi nitis wemi takwicken
sakima kichwon.
36. Kichitamak sakimanep wina-
kununda.
37, Wapahakey sakimanep shey-
abian.
n:
38. Elangomel sakimanep make-
liwulit.
39. Pitenumen sakimanep unchi-
<3) hillen.
tx^ r-^
40. Wonwihil wapekunchi wap-
sipayat.
L
4- 41. Makelomush sakimanep wu-
i latenamen.
THE WALAM OLUM. 211
\2. Again an Affable was chief, and made peace with all,
33. All were friends, all were united, under this great
chief
36. Great-Beaver was chief, remaining in Sassafras land.
37. White-Body was chief on the sea shore.
38. Peace-Maker was chief, friendly to all,
39. He-Makes-Mistakes was chief, hurriedly coming.
40. At this time whites came on the Eastern sea.
41. Much-Honored was chief; he was prosperous.
212 THE LENAPE AND THEIR LEGENDS.
42. Wulakeningus sakimanep
shawanipalat.
43. Otaliwako akowetako ashki-
palliton.
44. Wapagamoshki sakimanep
lamatanitis.
45. Wapashum sakimanep tale-
gawunkik.
^■^^^--fD^N 46. Mahiliniki mashawoniki ma-
konowiki.
47. Nitispayat sakimanep kipe-
mapekan,
Q — O 48. Wemiamik weminitik kiwik-
" "TX — hotan.
49. Pakimitzin sakimanep tawa-
nitip.
THE WALAM OLUM. 213
42. Well-Praised was chief; he fought at the south.
43. He fought in the land of the Talega and Koweta.
44. White-Otter was chief; a friend of the Talamatans.
45. White-Horn was chief; he went to the Talega,
46. To the Hilini, to the Shawnees, to the Kanawhas.
47, Coming-as-a-Friend was chief; he went to the Great
Lakes,
48, Visiting all his children, all his friends.
49. Cranberry-Eater was chief, friend of the Ottawas.
214 THE LENAPE AND THEIR LEGENDS.
50. Lowaponskan sakimanep
ganshowenik.
51. Tashawinso sakimanep shay-
abing.
52. Nakhagattamen nakhalissin
wenchikit,
52, bis. Unamini minsimini chi-
kimini.
53. Epallahchund sakimanep
mahongwipallat.
54. Langomuwi sakimanep ma-
hongwichamen.
dXi)
55. Wangomend sakimanep ika-
lawit,
56. Otaliwi wasiotowi shingalu-
sit.
THE WALAM OLUM. 215
50. North-Walker was chief; he made festivals.
51. Slow-Gatherer was chief at the shore.
52. As three were desired, three those were who grew
forth.
52. bis. The Unami, the Minsi, the Chikini.
53. Man-Who-Fails was chief; he fought the Mengwe.
54. He-is-Friendly was chief; he scared the Mengwe,
55. Saluted was chief; thither,
56. Over there, on the Scioto, he had foes.
216 THE LENAPE AND THEIR LEGENDS.
D:=-.
/^
57. Wapachikis sakimanep shay-
abinitis.
58. Nenachihat sakimanep pek-
linkwekin.
P^^"~-^,S____n 59- Wonwihil lowashawa wapay-
^ ^ achik.
zrf^ 60. Langomuwak kitohatewa
f^ . r- 7 ewenikiktit?
THE WALAM OLUM. 217
57. White-Crab was chief; a friend of the shore.
58. Watcher was chief; he looked toward the sea.
59. At this time, from north and south, the whites came.
60. They are peaceful ; they have great things ; who are
they?
NOTES.
j8®" The references to authorities on Algonkin picture-writing are the Appendix to
Tanner's Narrative of Captivity and Adventtires, Copway's Traditional History of
the Ojibmay Nation, and Schoolcraft's Synopsis of Indian Symbols, in Vol. I of his
History attd Statistics of the Indian Tribes. I have not pursued an investigation of
the symbols beyond the first chant.
I.
1. Rafinesque translates wemiguna "all sea-water." The proper
form is wemmguna, "at all times" (Anthony). The symbol is
that of the sky and clouds above the earth. Compare Copway,
p. 134; Schoolcraft, Synopsis, Fig. 17.
2. Kwelik, a dialectic form of quenek, Z. long, stretched out.
Kitanito, a compound of kehtan, great, and manito, mysterious
being, is rendered by Raf. as Creator ; wit is the substantive verb-
affix.
Heckewelder (MSS.) distinguishes between the synthetic form,
ketanittowit, which he translates " Majestic Being," and the
analytic form, kitschi matiito, which he renders " Supreme
Wonder-doer." In the latter, the sense of manito is brought out.
In the Delaware and related dialects it conveys the idea of making
or doing (jnaniton, to make, Zeisberger, Gram., p. 222 ; maranito
taendo, make a fire, Campanius ; Chipeway, win ina-nitawito , he
himself makes it, or, can make it).
The idea of making or creating is at the bottom of many native
titles to supernatural powers, as the Shawnee We-shellaqua, "he
that made us all." (Rev. David Jones, Journal of Two Visits,
etc., p. 62.) See notes to line four. The Algonkin root, etu, he
does, he acts, he makes, would therefore seem to be a radical of
the word. (See Howse, Gram, of the Cree Lattg., p. 160.)
Dr. Trumbull, on the other hand, believes the only radical to be
an, = el ox al,m the sense of "to be more than," "to surpass," "to
exceed;" and maintains that the syllable it, of the theme manit,
is a formative suffix. (In Old and New, March, 1870.)
Heckewelder, in his translation "wonder-doer," recognizes the
219
220 THE LENAPE AND THEIR LEGENDS.
force of both elements, and from the analogous expressions I have
quoted, is probably correct. The element an is thus an intensive
prefix to the real root //, and the compound radical thus formed in
the third person, singular, inanito, means "he or it does or acts in
a surpassing or extraordinary manner."
Essop, pi. essopak, frequently recurring words, are suppositive
(see p. 90) forms of the verb lissm, "to be or do so, to be so
situated, disposed, or acting" (Zeisberger, Gram. p. 117). The
terminal p is the sign of the preterite. They are dialectic for
elsitup and elsichtittip.
The symbol of a head with rays represents a manito. School-
craft, Synopsis, Fig. 10.
3. Squier omits the word ebimamek. These terms are formal
epithets applied to the highest divinity. See page 158.
Squier also adds that Fig. 3 represents the sun, and is the symbol
of the Great Spirit. Both these statements are incorrect. The oval
is the earth-plain, with its four cardinal points, and the dot in the
centre signifies the spirit. See Copway, p. 135.
4. Sohalawak is not a Delaware form, but is a true Algonkin
word, as seen in the Cree ooseh-ayoo, animate, ooseh-taw, inani-
mate, he, it, makes, produces. (Howse, Cree Grammar, p. 166.)
It appears in the Shawnee w' shellaqua, quoted in notes to verse 2 ;
in the Minsi dialect the corresponding word is kwishelmaivak ;
owak is a mistake for ivoak, and Rafinesque translates it " much
air." Awasagamak, heaven, sky, literally, " the land or place
beyond," from awossi, beyond ; but Dr. Trumbull prefers a deriva-
tion from a root signifiying " light," Del. waseleu, it is clear or
bright (Trans. Am. Philol. Soc, 1872, p. 164); this latter appears
to me overstrained. The symbol is the earth surmounted by the
sky.
5. The symbol represents the sun, moon and stars in the sky,
which is repeated with change of relative positions in the next
verse. In Minsi, the fifth line would read, Kwishelmaivak kischohk
nipahenk alankwewak.
7. On the termination wagan see page loi. The prefix ksh,
properly k'sch, is intensive, as it is an abbreviation of kitschi,
great, large. Thus sokelan, it rains, k'schilan, it rains very hard.
The symbol seems to indicate the waters flowing off.
8. Mr. Anthony renders this line in Minsi : —
NOTES. 221
Pilikin ameni-menayen epit,
Grew-clean groups of islands where they are;
That is, that the islands rose dry and clean from the water, as
they now are found.
Delsifi-epit ; the first part of this compound, properly lu'deil-
sinewo, is the indicative present, 3d p. pL, of lissin, to be thus, or
so situated ; epit is what Zeisberger [Gram. p. 115) calls the "ad-
verbial " form of achpin, to be there, in a particular place. This
adverbial is really the suppositive form of the verb, after the vowel-
change has taken place. (See above, page 107.)
Former renderings of the line are : " It looks bright, and islands
stood there " (Rafinesque). "All was made bright, and the islands
were brought into being " (Squier).
The symbol is a three cornered point of land, rising above the
water under the sky.
9. Manito manitoak, "made the makers'," Raf. ; "made the
Great Spirits," Squier. Either of these renderings is defensible,
as will appear from the senses of manito, above given.
This line can be read in Minsi, Lapi-iip Kehtanitowit man'ito
inanVtowak, Again-he-spake, Great-Spirit, a spirit, spirits. The
symbol represents the communion of the spirits. Compare Tan-
ner, Narrative , p. 359, fig. 24.
10. Raf. and Squier absurdly translate angelatawiwak, angels.
It is from a familiar Del. verb, angeln, to die. Compare Abnaki
8anangmes8ak, " revenants," Rasles, and w' tanglotvagan, his
death, Zeis. The form in the text, according to Mr. Anthony, has
the sense, "things destined to die," mortal, perishable. He
gives the line in Minsi as follows : —
Aweniwak angelatawaivak wtschitschivaiikwak wemiwak.
Beings mortals souls and all.
The wak of the last word is not the plural but the conjunction
"and ;" as in the Latin, omniaqne.
11. Raf. translates //« WW as "man-being," and Squier thinks it
the Chipeway inini, men ; but it appears to be the adverb janwi,
ever, always. The symbol is apparently that of birth, or being
born. Compare Tanner, Narr., p. 351, fig. i, with that meaning,
an armless figure with wide spread legs.
12. The pictograph is a woman, with breasts, but armless. The
" first mother" here represented was an important personage in
222 THE LENAPE AND THEIR LEGENDS.
the 'mythology of the Chipeways and neighboring tribes. She was
called "the grandmother of mankind " {^Me-suk-kum-me-go-kwa,
in Dr. James' orthography), and it was to her that Nanabush
(Manibozho), imparted the secrets of all roots, herbs and plants.
Hence, the medicine men direct their songs and addresses to her
whenever they take anything from the earth which is to be used as
a medicine. Tanner's Narrative, p. 355.
13. The figure of a square, the world, with the four varieties of
animals named.
14. The bad spirit was, in Algonkin mythology, the water god,
and was represented as a serpent-like figure. See Copway, pp.
134, 135. Schoolcraft, Synopsis, figs. 93, 100.
Amang-amek, plural form of the compound amangi, great ;
namaes fish ; but amangi has the associate idea of terrifying,
frightful, hence the reference is to some mythical water monster
(Cree, am, faire peur, Lacombe).
Raf. translates both nakowak in this line, and 7iakowa, in II,
6, as "black snake." They can have no such meaning, black,
in Lenape, being suckeu, and in none of the Algonkin dialects
does nak mean black.
16. The figure represents the earth-plain under the form of the
area of a lodge, with central fire and the people in it, typifying
friendliness. Comp. Tanner, Narr., p. 348, fig. i.
V. 16 pursues the topic of v. '13, and it looks as if v. 14 and 15
should be transposed to follow v. 20.
17. The former renderings are . —
" Thou being Kiwis, good God Wunand, and the good makers
were such." — Rafinesque.
"There being a good god, all spirits were good." — Squier.
Rafinesque mistook the adverb kiiuis for a proper name.
18. Raf. translates nijini, the Jins, and nantinezuak, fairies, and
Squier follows him in the latter, but could not go as far as the
former ! As seen in the vocabulary, I attach wholly different
notions to these words. The two figures united refer to the sexual
relation. Compare Tanner, Narr., pp. 371, figs. 8, 9.
19. Gattamin cannot mean "fat fruit," as Raf. translates it.
He has evidently mistaken the explanation given by Heckewelder,
of Catawissa, Gattawisit, becoming fat, and thought that gatta,
was fat, whereas wisu is " fat." (Zeis. Gram., p. 229.)
NOTES. 223
H'akon is understood by Rafinesque as the proper name of the
evil spirit, connecting it with the Dakota wakan, divine, super-
natural.
20. The dream of "the good old times," the happy epoch
of yore, when men dwelt in peace and prosperity, was, as I
have shown, page 135, a myth of the Delawares, and George
Copway tells us that the Chipeway legends also recalled it with
delight. ( Traditional History of the Ojibway Natio7t, pp. 98 and
169-175.)
21. The symbol is the same as that of the " bad spirit under the
earth," given by Copway, p. 135.
A similar figure is given by Copway to signify "bad," p. 135.
I do not understand its allusion.
22. Mattalogas ; the prefix is the negative matta, no, not, and
generally conveys a bad sense, as niatteleman, to despise one,
mattelendani, to be uneasy. Zeis.
Pallalogasin, to sin, from palli, elsewhere, other than, hence
palUiiken, to shoot amiss, to miss the mark, to go wrong.
Maktaton, unhappiness. There is a relation in Lenape between
the negative matta, in Minsi, machta, and the words for bad, ugly,
evil, and the like ; niachtisisu, here it is bad, or ugly. Zeisb. It
would seem to be an intuitive recognition of the profound philo-
sophical maxim that evil is ever a negation ; that Mephistopheles
is, as he says in Faust —
" Der Geist der stets verneint."
23. The symbol is apparently trees on hills, bent by a storm,
and beneatt^ a death's head.
24. The picture seems to be two countries connected by a
bridge.
Atak kitahican, = attach, beyond, above ; kitahican, the ocean,
literally "the great tidal sea." It is possible this has reference to
the deluge, which is described in the next section; but usually
kitahican meant the ocean.
II.
I. Maskanako ; the Lenape words would be inechek, great, ach-
gook, snake ; but maska is more allied to the Cree inaskaw, strong,
hard, solid. Raf, translates the close of the line " when men had
become bad."
224 THE LENAPE AND THEIR LEGENDS.
2. Schingalan, to hate ; from the adjective schingi, dishking;
unwilling. This is the contrary of wingi, liking, willing. Both
are from the subjective radical n or ni, I, Ego, the latter with the
prefix wel, signifying pleasurable sensation (see page 104).
Shawelendainep , preterite form, strengthened by the prefix ksch,
of the verb acquiwelendam, Zeis., to disquiet, to trouble; it has not
the passive sense given in Rafinesque's translation. All verbs
terminating in eletidam signify a disposition of mind, the root
being again the subjective 11, ego. Raf. translates: "This strong
snake had become the foe of the Jins, and they became troubled,
hating each other."
3. Palliton, from palli, elsewhere (from what was intended),
hence "to spoil something, to do it wrong," and later "to fall out,
to fight."
Lungundowin, from langan, easy, light to do, Chipeway, 7tin
nangan, I find it light, of no trouble ; hence, "peace," as being a
time free from trouble ; and by a third application of the idea,
elangotJiellan, friends, those who are at peace with us.
4. Raf. translates this line : " Less men with dead-keeper fight-
ing," which is a total misunderstanding of the words. On the
derivation of nihanlowit seo. ante, page 102.
6. On nakowa, see I, line 14. Here I consider it a derivative
from nacha, three, and both the sense of the line and the symbol,
with three marks to the right of the figure, indicate this meaning.
The three antagonists are the monster, the waters, and the Great
Snake himself.
7. The repetition of the words is to add force to the phrase.
8. This is an important line, as indicating the origin of the
Walam Olum. Nanaboush is not the Delaware form of the name
of the Algonkin hero-god, so far as known, but the Chipeway
NanaboosJwo, Tanner, Nanibajoti, McKinney, properly Ndnaboj,
the Trickster, the Cheater, allied to Chip, nin nanabanis, I am
cheated. This term, like the Cree Wisakketjak, which has the
same meaning [fow-be, trompeitr, Lacombe), was applied to the
hero-god of these nations on account of his exhaustless ingenuity
in devising tricks, ruses, disguises and transformations, to over-
come the various other divine powers with whom he came in con-
flict. This seemingly depreciatory term arose from the same
admiration of versatility of powers which has imparted such uni-
NOTES. 225
versal popularity to the story of the wily (jxoXuTpoT:o<s) Ulysses,
and the trickery of Master Reynard.
The appearance of this form of the name indicates that the
version of the legend here given has been influenced byChipeway
associations, as, indeed, we might expect, since it was obtained in
Indiana, where the Delawares were in constant intercourse with
their Chipeway neighbors.
Tulapit nienapit^tulpe epit, menatey epit, "it was then at the
turtle, it was then at the island." The form Tula has given rise to
the strangest theorizing about this line, as, of course, the antiquaries
could not resist the temptation to see in it a reference to the Tula
or Tollan of Aztec mythology, the capital city of the Toltecs and
the home of Quetzalcoatl.
The similarity of the words is purely fortuitous. The Lenape
word tiilpe means turtle or tortoise, especially, says Zeisberger, a^
water or sea turtle. In their mythology, as I have already shown
(ante, p. 134) the earth was supposed to be floating on a bound-
less ocean, as a turtle floats on the surface of a pond. Hence,
symbolically, the turtle represents the dry land.
Maskaboush = Chip, mas/ika, strong, wabos, usually translated
hare or rabbit, but really "White One." I have fully explained
this mistaken sense of the word in American Hero Myths, pp. 41,
42, and elsewhere.
9. The Algonkin myth relates that Michabo or Nanaboj after
having formed the earth on the primal ocean, walked round and
round it, and by this act increased it constantly in size.
Rafinesque's translation is: — "Being born creeping, he is ready
to move and dwell at Tula; " and in his note to the line he adds,
"Tula is the ancient seat of the Toltecas and Mexican nations in
Asia; the Tulan or Turan of Central Tartary."
The entire absence of connected meaning in this and other lines
of Rafinesque's translation is strong evidence that he did not fabri-
cate the text; otherwise he would certainly have assigned it some
coherent sense.
The turtle is, as usual, the symbol of the land or earth (see
page 133).
12. Manito-dasin, the Divine Maiden, or the Daughter of the
Gods, as it might be freely translated. The reference is to the
Virgin who at the beginning of things descended from heaven, and
226 THE LENAPE AND THEIR LEGENDS.
alighting on the back of the turtle became the mother of Nanaboj
and his brothers. She was well known in Eastern Algonkin
mythology, as I have already shown. (See above, p. 131.)
13. This and the three following verses form, observes Rafin-
esque, a rhymed hymn to Nanabush.
14. In this line the men are referred to as Linapi, not lennowak
as before. Here then begins the particular history of the Lenape
tribe, whose chief sub-tribe was the Turtle clan.
The meaning of the line is very obscure. It seems to refer to
the origin of the Unami, or Turtle sub-tribe of the Delawares.
16. Kwamipokho, translated by Raf. "plain and mountain,"
does not appear to me to bear any such rendering. I take it as a
form of chavipeecheneii, Z. " it is still or stagnant water," the
appropriateness of which to the context is evident.
Sitwalikho, Raf. renders "path of cave," deriving it obviously
from tsit, foot, and tvoalhett, a hole. It has no sort of meaning
in this rendering, and I assume, therefore, that it is a derivative
from tschiiqiii, silent.
Maskaji luagan, probably an error for viaskanakon, as in v. i.
Palliwi, palliwi, "is elsewhere, is elsewhere," or, "is foiled, is
overcome."
III.
I. Wittmik talli : in the MS. these words are first translated
" dwelling town there," but the last two words are erased and " of
Talli" substituted. This is one of a number of instances where
Rafinesque altered his first translations, which is further evidence
that he did not manufacture the text. In this instance, as fre-
quently, he altered it'for the worse. Wittank is from witen, to go
with or be with, Zeis., and talli is the adverb " there."
3. ]\Ieshaittang, "many deer" (see Vocabulary), translated by
Rafinesque, "game."
Silieiuak, rendered by Rafinesque silt, cattle, ewak, they go.
The wak is the terminal "and" (see notes to I. v. 10). The
word ^zjrz7^, in modern Delaware j-zV//^/^ (Whipple's Vocabulary),
means "buffalo." Its older form is seen in the MS. vocab. of
the New Jersey Indians, 1792, where it is sisiliamnus. This is
a compound of the generic termination vmus, Cree, viustus
(whence our word "moose"), meaning any large quadruped, and
NOTES. 227
probably the prefix tschilani, strong, powerful, with an intensive
reduplication.
4. Poivalessin, from the same root as poivzvoiv (see page "jo).
The course of thought was that the dreamer {powwow) became
wise beyond his followers, and hence obtained power and riches,
though not of a martial character.
Elowichik, hunters ; allowin, to hunt, doubtless connected with
alliins, an arrow.
5, 6. A note in the MS. states that the symbols of these two
verses were united together in the original drawings.
7. In this verse the pre-eminence of the Turtle sub-tribe, the
Unami, is asserted to have obtained from the most ancient times.
8. The verses 8, 9, 10, are referred in Rafinesque's free trans-
lation to the " Snake people." They seem to me to be descriptive
of the grief of the Lenape on leaving their ancient home.
12. PokhapokJiapek, "Gaping Sea," Raf. Both this and the
preceding word are descriptive of the sea, referred to as offering
means of subsistence ; namaes, fish, pocqueu, muscles or clams,
being the two main food-products of the water for the Indians.
The location of this productive spot I leave for future investi-
gators to determine. The Detroit River and the Thousand Isles
in the St. Lawrence are the most appropriate localities, to my
mind.
13. The last word of the line is given in the MS. both as
vienakinep, and ako7nenep, the latter a later interlineation. I pre-
fer the former.
Wapasinep, may mean " at the East," as well as " in the light."
The latter is a metaphor, common in the native tongues, for pros-
perity.
Verses 13 to 20 inclusive were printed by Rafinesque in the origi-
nal, and called by him, " the poem on the passage to America," as
he understood this narrative to refer to the period when the an-
cestors of the Lenape crossed Behring straits from Asia to America
on the ice.
17. Kitahican. This is the term given by Zeisberger to the
Ocean. The prefix Kit is "great," and the termination hican
appears to have been confined to tidal waters (see above, p. 21).
Elsewhere this termination signifies an instrument. Probably it
was applicable to all large bodies of water.
228 THE LENAPE AND THEIR LEGENDS.
On pokhakhopek, doubtless a carelessness for pokhapokhapek,
line 12, see note to the latter.
l8. Squier does not give the numerals, but says simply " in vast
numbers." No doubt this is the intention of the expression.
20. Shiwaking, "the place of spruce firs " (see Vocab).
They crossed in mid-winter a broad stream, rich in fish and
shell-fish, and arrived at a land covered with forests of spruce.
For a long time this appears to have remained their home.
IV.
2. Sittamaganat , Raf. translates "Path Leader." The word
tamagaiiat appears in other verses, as w'iainaganat, IV, 37 ;
iamaganat, IV, 55 ; taniaganend, V, 2. I derive it from the root
tmn, literally to drink, but generally, to smoke tobacco, as in
Roger Williams' Key wut-tamjtiagofi, a pipe (see above, page 49).
Hence I take tamagatnat to be the pipe-bearer, he who had charge
of the Sacred Calumet. If it is objected that this puts the use of
tobacco by the Lenape too remote, I reply that we do not know
when they began to use it, and moreover, this may be an anach-
ronism of tradition.
13, 14. The lands toward the four cardinal points are described
from a centre where the tribe was then located. Neither Rafi-
nescjue nor Squier understood this, and their renderings do not
mention the territories North and West. From the description, I
should place the then location of the tribe in Western New York
and Northern Ohio.
16. The four names seem to be appellatives of different tribes.
One of the extinct tribes remembered in Chipeway tradition was
the Assigimaik, Stone People (Schoolcraft, History and Statistics
of the Ind. Tribes, Vol. I, p. 305).
25. The legend here relates that the cultivation of maize began
after they had reached a low latitude, presumably Southern In-
diana or Ohio. The legend of the New England Indians was
that a crow flew down from the great God Kitantowit, bringing in
one ear a grain of corn, in the other a bean, and taught them the
cultivation of these plants. (Roger Williams, Key into the Lan-
guage of America, p. 114.) See further, ante, p. 48.
34. Wisawana, the Yellow River. There is a small river, so-
called, in the State of Indiana, a branch of the Kankakee, called
NOTES. 229
on Hough's "Map of the Indian Names of Indiana" We-tJio-gan,
a corruption of wisawanna. (See Hough's map, in Twelfth An-
nual Report of the Geology and Natural History of Indiana,
1883.) When the Minsi made their first migration west, about
1690, they directed their course to this spot, where they were found
by Charlevoix in 1721.
36, Tamenend, the name of the celebrated chief now better
known to us as Tammany, who dealt with William Penn. Hecke-
welder translates it as "Affable." This is the first of the name.
A second is mentioned, V, 32. The friend of Penn was the third.
46. Towakon pallitonep, Raf. translates " father snake, he was
mad ! "
48. Perhaps this line should be translated : " They speak well
of the east ; many go to the east."
49. Nemassipi, Fish River. In the MS. this name was first
written mixtu sipi. The name "Fish River" was applied to
various streams by the Delawares, but never, so far as I know, to
the Mississippi. In the present connection it seems to refer either
to the St. Lawrence, about the Thousand Isles, or else its upper
stream, the Detroit River, both of which were famous fishing
spots.
50. Talligewi. No name in the Lenape legends has given rise
to more extensive discussion than this. It is usually connected
with Alligewi and this again with Alleghany. This seems sup-
ported by Loskiel, who, writing on the authority of Zeisberger,
says, " Nun nennen die Delawaren die ganze Gegend, so weit die
Gewasser reichen, die in dem Ohio fallen, AUigewinengk, welches
so viel bedeutet, als das Land, in welches sie sich aus weit ent-
fernten Orten begeben haben." {Geschichte der Mission, etc., p.
164.)
The meaning here assigned to AUigewinengk, " land where they
arrived from distant places," is evidently based on the resolution
of the compound into talli, there, icku, to that place, ewak, they
go, with a locative final. The initial / is often omitted in adverbial
compounds of talli (itself a compound oi ta, locative particle, and
//, to), as allamu7ik, in there.
Bishop Ettwein gives to the word a different meaning. He
writes : " The Delawares call the western country Alligeiuetiurk,
which signifies a War-Path ; the river itself they call Alligewi
230 THE LENAPE AND THEIR LEGENDS.
Sipo." [Legends and Traditions, etc., in Bull, of the Pa. Hist,
Soc. p. 34.) Here the derivation would be from palliton, to fight,
ewak, they go, and a locative, "they go there to fight." The
omission of the initial p was not uncommon, as Campanius gives
ayuta =aliiton, to make war. {CatecJiisinus, p. 141.)
Basing his opinion on an expression in the Journal of C. F.
Post, to the effect that Alleghany means "fine or fair river," Dr.
J. H. Trumbull analyzes it into wulik, hanne, sipu, which he trans-
lates "best, rapid-stream, long-river [Connect. Hist. Soc. Colls,
Vol. II).
Rafinesque, in the MS. of the Walum Olum, gives Talligewi the
translation "there found," from talli, there, and I know not what
word for " found."
There have not been wanting those who would derive the name
Alleghany from Iroquois roots, as the Seneca De-o-na-ga-no,
" cold water" [Amer. Hist. Mag. Vol. IV, p. 184). But there is
no probability that the word is Iroquois,
Whatever its origin, the name was not confined to the Alle-
ghany river, but included the whole of the Upper Ohio, as the
interpreter Post distinctly says.
The Rev. Mr. Heckewelder was of opinion that Talligewi
was a word foreign to the Algonkin, a noinen gentile of another
tribe, adopted by the Delawares, just as they adopted Mcngwe for
the Iroquois from the Onondaga Yenkwe, men (see above, page
14). It is not necessarily connected with Alleghany, which may
be pure Algonkin. He says, " Those people called themselves
Talligeu or Talligewi.'' [Lidian Nations p. 48.) The accent, as
hegives it, Tallige'wi, shows that the word is, Tallike, with the sub-
stantive verb termination, so that Talligewi va.QZ.ns," He is a Tal-
like," or, " It is of (belongs to) the Tallike."
This appears to me the most probable supposition of any I have
quoted, and it reduces our quest to that of a nation who called
themselves by a name which, to Lenape ears, would sound like
Tallike. Such a nation presents itself at once in the Cherokees,
who call themselves Tsa'laki. Moreover, they fill the require-
ments in other particulars. Their ancient traditions assign them a
residence precisely where the Delaware legends locate the Tallike,
to wit, on the upper waters of the Ohio (see above, page 17).
Fragments of them continued there until within the historic
' NOTES. 231
period; and the persistent hostility between them and the Dela-
wares points to some ancient and important contest.
Name, location and legends, therefore, combine to identify the
Cherokees or Tsalaki with the Tallike ; and this is as much evi-
dence as we can expect to produce in such researches. I can see
no reason whatever for Dr. Shea's opinion that the Lenape " in
their progress eastward drove out of Ohio the Ouappas, called by
the Algonkins, Alkanzas or Alligewi, who retreated down the
Ohio and Mississippi," (Shea, Notes to Alsop's Maryland, p.
ii8.)
The question remains, whether the Tallike were the " Mound
Builders." It is not so stated in the Walam Olum. The infer-
ence rather is that the " Snake people," Akowini ox Akonapi, dwelt
in the river valleys north of the Ohio river, in the area of Western
Ohio and Indiana, where the most important earth-works are
found — and singularly enough none more remarkable than the
immense effigy of the serpent in Adams County, Ohio, which winds
its gigantic coils over 700 feet in length on the summit of a bold
bluff overlooking Brush Creek.
According to the Red Score, the Snake people were conquered
by the Algonkins long before the contest with the Tallike began.
These latter lay between the position then occupied by the Lenape
and the eastern territory where they were found by the whites.
In other words, the Tallike were on the Upper Ohio and its tribu-
taries, and they had to be driven south before the path across the
mountains was open. . For this reason they are called wapawul-
laton, "possessing the East," that is, with reference to the then
position of the Lenape in southwestern Ohio.
54. Talaniatan. This was the Lenape name of the Huron-
Iroquois or Wyandots. It is found in the form Telamatinos in a
"List of II Nations living West of Allegheny" present by deputy
at a Conference in Philadelphia, 1759 {Minutes of the Prov.Coiincil
of Penna., Vol. VIII, p. 418). Heckewelder gives Delainatterios
[Ind. Nations, p. 80).
Rafinesque translates the name in one place by " not Talas,"
and in another by " not of us," from Len. matta, not, Latin 7tos,
us. That the Lenape did not speak Latin made no difference in
his linguistic theory, as he held all languages to be at core the
same ! On the Hurons, see above, p. 16.
232 THE LENAPE AND THEIR LEGENDS.
V.
2. Wapala7ieng, apparently the White River, Indiana, or else
the Wabash.
l6. In this line the three tribes are mentioned which were previ-
ously named in IV, 44, 45, 46, and the difference in the spelling
shows that the chant was written down by one unacquainted with
the forms of the language. The correspondent names are : —
IV. V.
Akowini, Sinako.
Towakon, ■ Towako.
Lowanuski, Lowako.
The termination ako, uniformly rendered by Rafinesque sjiake,
appears to be either the animate plural in ak, or the locative aki,
place or land.
The Towako are probably the Ot-tawa called by the Delaware
Taivay; or the Twightees, called by them Tawatawee (see "List
of II Nations," etc., in Minutes of the Prov. Council of Pa., Vol.
VIII, p. 418).
There is difficulty in reconciling Akowini and Sitiako. In the
former, the prefix ako may be from achgook, snake, as Rafinesque
and Squier rendered it.
The yjord Lowa?tuski appears again in v. 18, where Raf. inserts
the note, " Lowushkis are Esquimaux." It means simply "winter
land," or " Northern people," and is not likely to have any refer-
ence to the Eskimo.
22. "Without snakes," z. e., free from enemies.
24. On the derivation of Susquehannah, see page 14.
25. Winakakifig, Sassafras Land, the native name of eastern
Pennsylvania.
29. The Wapings and the Minsi seem to be referred to.
33, 36. The omission of the numbers 34 and 35 is in the
original MS.
50. Ganshowe7iik ; Raf. translates this "the noisy place, or
Niagara." It is a derivative from the root kan. See Vocab.
60. Ewenikiktit, may be translated "whites" or "Europeans."
See Vocabulary,
VOCABULARY.
In the following Vocabulary the meaning placed immediately after the word is that
assigned to it in Rafinesque's original MS.; the probable composition of it is then
added, with its correct rendering. The standard of the language adopted is that of
the Moravian missionaries (see above, p. gy). The initials referring to authorities
are: Z., for Zeisberger, K., for Kampman, H., for Heckewelder, R. W., Roger
Williams, C. or Camp., Campanius, etc.
Aan. I, 6. To move ; to go ; Z. conjugated, Gram., p. 142. Chip, ani,
he goes ; aunj-eh, he moves. Cf. Payat.
Agamunk. Ill, 16. Over water. Acawenuck, over the water. R. W.
Acawmenoakit, land on the other side of the water, i. e. England.
R. W. The proper names Accomac, Algonkin, etc., are from the
same roots.
Agunouken. Ill, 13. Always our fathers. Nooch, my father, Z. in
which n is the possessive our or my.
Akhokink. Ill, 9. Snake land at. Derivatives beginning with akho,
and some with ako appear to be compounds of achgook, Mohegan
ukkok, the generic name for snake.
Akhomenis. IV, 3. Snake Island. Menaiey, island, and achgook,
snake.
Akhonapi. IV, 16. Snaking man. Achgook, zxi^ a/^, man; & nomen
gentile.
Akhopayat. IV, 6. Snake coming. Achgook, sn2.ViQ; pay a ifYit comts.
Akhopokho. IV, 6. Snake hill. Achgook, snake. Pockhepokink, a
river between hills. Heck.
Akhowemi. IV, 7. Snake all. Achgook, snake, and wemi, all.
Ako. II, I, 2. Snake. Achgook, snake. See Akhokink.
Akolaki. IV, 13, and Akolaking. IV, 18. At beautiful land. Ach-
gook, snake ; aki, land. A form of Akhokink, q. v.
Akomen. Ill, 14, 18. Island snake, Achgook, snake; menatey,
island.
Akomenaki. Ill, 10. Snake fortified island. Akomen, q. v., and aki,
land.
Akomenep. Ill, 13. Snake island was, Akomen, with the preterit
termination.
Akopehella. II, 6. Snake water rushing. Kschippehellan, strong
stream in a river, Z. See Pehella.
p 233
234 THE LENAPE AND THEIR LEGENDS.
Akowetako. V, 43. Coweta snakes. JVeia, a house, H., and aii,
land ; the Coweta land.
Akowini. IV, 44. Snake beings or like. The Snake people; a nomen
gentile.
Akpinep. Ill, 2. Was there. Achpil, to stay, abide; achpiney, a
sleeping place.
Alankwak. I, 5. Stars. Alank, star.
Alkosohit. IV, 26. Keeper and preserver. Allouchsit, strong and
mighty. K.
AUendyachick. IV, 32. Some going. Alende, some.
Allendhilla, IV, 52. Some kill, ^/^zo't', some, and «?7^?7/a«, to kill.
Allendyumek. II, 1 1. Some of them.
Allowelendam. Ill, 20. Preferring above all. Allowelendamen, to
esteem highly. Z.
Allumapi. 111,19. With dogs of man. AllumjAog; ape,mz.n; men
having dogs.
Alokuwi. IV, 46. Lean he. Alociiwoagan, leanness. Z.
Amangaki. V, 21. Large land. Amatigi.^ great, large. See p. 146,
note.
Amangam. II, 6. Monster. Aviangi. See p. 146, note.
Amangamek. I, 14. Manitos or large reptiles. II, li. Waters of
sea. Amangemek, a large fish.
Amokolen. Ill, 13. Boating. Amochol, canoe or boat.
Amigaki. V, 21. Long land. Afuangi, great; a/a', land.
Angelotawiwak. I, 10. Angels also. From angebi, to die. See note
to the passage.
Angomelchik. IV, 4. The friends or friendly souls. Jlfclec/iiischajtt,
soul. Z. ; inelih, corruption, Z., and angeln, to die ; " the souls de-
parted."
Anup. II, I. When. /^rt«?</, when or if I went. Zeis. 6^;-d'OT.,p. 143.
Doubtful.
Apakachik. Ill, 6. Spreaders. Apach ischiechlon, to display, to attach
oneself to or upon. K.
Apakchikton. IV, 11. Spreading. Stt Apakachik.
Apendawi. IV, 26. Useful he. Apendamen, to make use of; apensuwi,
useful, enjoyable.
Aptelendam. HI, 9. Grieving. To grieve to death. Zeis.
Askipalliton. V, 43. Must make war. Aski, must, obliged, and pal-
liton.
AskiwaaL IV. They must go. Aski, must, and aan or aal, to go.
Assinapi. IV, 16. Stone man. y^j-j-w, a stone ; s/^, a man; a nomen
gentile.
Atak. I, 24. Beyond. Attach, beyond, above. Zeis.
Atam. HI, 8. Let us go. Atam, let us go. Z. Gram.
VOCABULARY. 235
Attagatta. IV, 31. Unwilling. AUa, or ma^^a, negative -prefix; ga^ta,
to want, or wish.
Attalchinitis. IV, 62. Not always friend. AUa, neg. prefix ; niiap,
friend, or our friend.
Attaminin. IV, 28. No corn. yi//a', neg. prefix ; w^m, berry or corn.
Attasokelan. IV, 28. No raining. Atta, neg. prefix ; sokeian, rain.
Awasagamek. I, 4. Much heaven. Awosegame, heaven. Z.
Awesik. I, 13. Beasts. Awessis, a beast.
Awolagan. V, 12. Heavenly. Awullakeiiim, to praise. K.
Ayamak. IV, 15, 17. The great warrior. ^y«;«;«^«, to buy, purchase.
K. ; from aji, take it! hence "the Buyer," or "the Seizer."
Chanelendam. Ill, 20. Doubting. Tschannelendam, to consider, to
be in doubt. K.
Chichankwak. I, 10. Souls also. Tschitschank, soul.
Chihillen. Ill, 11. Separating. Tschitschpihieleii, to split asunder;
cf. chipeu, it separates.
Chikimini. V, 52. Turkey tribe. See above, p. 37.
Chikonapi. IV, 16. Robbing man. Cheche, to rob, R. W., Key, p.
102.
Chiksit. Ill, 5. Holy. Kschiechek, clean; kschiechmichsopannik,
holy. Z.
Chilili. IV, 10, 12, 15. Snow-bird. Chilili, snow-bird, Heck. Ind.
Names, p. 363.
Chingalsuwi. IV, 30. Stiffened he. Tschingahti, stiff.
Chintanes. Ill, 4. Strong. Tsckintamen, strong. Z.
Chitanesit. Ill, 5. Strong. Tschiiani, strong. K.
Chitanitis. IV, 51. Strong friend. Tschitani^^ixovi'g; wz'/zV, friend.
Chitanvs^ulit. IV, 45. Strong and good. Tschitani, strong; lunlit.
Cholensak. I, 13. Birds. Tscholens, bird.
Dasin. II, 12. Daughter. N^danilss, my daughter.
Danisapi. 111,19. Daughters of man. .A^V««/m, my daughter ; ape,
man.
Delsin. I, 8. Is there. WdeUsin, he is or does so. Zeis. Gram.,
p. 117.
Delsinewo. HI, 5. They are. Wdelhinewo, they are or do so. Zeis.
Gratn., p. 117.
Eken. II, 2. Together. Probably an error for nekama, those.
Elangomel, V, 38. Friendly to all. Elango7nellan, my friend. Z.
Elemamik. I, 3. Everywhere, Elemamek, everywhere. Z.
236 THE LENAPE AND THEIR LEGENDS.
Elendamep. I, 20. Thinking. On elendam, see above, p. 100.
Eli. 1,21. While. ^//, because, then, so, that. K. Also a superlative
prefix, as eli kimi, very privately.
Elmusichik. IV, 4. The goers. Elemussit, he who goes away. Z.
Elowaki. Ill, 17. Hunting country. Eluwak, most powerful. Z. In
this word and in elozvapi, Rafinesque mistook the meaning of the
prefix. Compare elowichik.
Elowapi. Ill, 19. Hunting manly. Eli, intensive, best or most, and
ape, man, or perhaps wapi, knowing.
Elowichik. Ill, 4, 5, 6. Hunters. From allauwin, to hunt. Z. ; allau-
witaa, let us go hunting. H.
Eluwi. Ill, 5. Most. The superlative form eli, with the substantive
verb suffix, wi.
Eluwiwulit. IV, 36. The best. From eluwi, and wulit, good.
Enolowrin. IV, 9. Things who. Doubtful, perhaps, nantte, those;
owini, beings, people.
Epallahchund. V, 53. Failer, who fails. Pallikiken, \.o ^oo'i ^xm%%;
palliaan, to go away.
Epit. I, 8. Being there. I, 24. At. This is a suppositive fonn from
achpiii, called the "adverbial" by Zeis., Gram., p. 115, who trans-
lates it " where he is." It may also be translated by the preposition
"at." See Heckewelder, Correspondence zoith Duponceau, Letter
XXI.
Eshohok. II, 7. Much penetrate. Eschoochwen, to go through. Z.
Essop. I, 2, 3. He was.
Essopak. 1,17. Were. II, l, 2. Had become. A form from /wj-z«,
to be or do so.
Ewak. Ill, 3. They go. Ewak, they go. Z. ; from aaji, to go.
Ewenikiktit. V, 60. Who are they ? Auwenik, who are they ? Z. Gram.,
116. The term ^7wz«?c^j was that applied to the whites in general
by the New England Indians. The Abbd Maurault derives it from
aSeni, who, uji, whence; = whence come they? Histoire des Abena-
kis, p. 10.
Gahani. II, 10. Shallow water. Gahan, shallow. K.
Gaho. I, 12. Mother. See Nigoha.
Gandhaton. IV, 7. Concealing or hiding themselves. Gandhatton,
to hide, to conceal. K.
Ganshowenik. V, 50. Noisy place (Niagara). Ganschewen, to roar,
to make a great noise, Z. ; or from kattti. See above, p. 73.
Gattamin. 1,19. Fat fruits. A^'^a/^aw^'w, I wish, desire. Z. See note
to passage.
Gattawisi. V, 25. Becoming fat. Gatla, do you want ? Z. ; gattawisi,
becoming fat, proper form of Catawissa. Heck., Ind. Names, p. 360.
See note.
VOCABULARY. 237
Gentikalanep. IV, 39. Festivals he made. Kanti, to sing and dance.
See p. 73.
Gichi. II, 5. Ready. See the root kick, p. 102.
Gikenopalat. V, 23. Great warrior. Gischigin, to be born; netopa-
lisak ^= warrior. Z.
Gishelendam. IV, 62. Conspiring. Gischelendam, to hatch or medi-
tate something good or bad. See p. 103.
Gishikin. II, 9. Being born. Gischigin, to be born. See pp. 102-3.
Gishikshawipek, V, 26. Sun salt sea. Gischihan, to make ; schejek,
wampum.
Gishuk. I, 5. Sun. See p. 103.
Gotatamen. IV, 51. He desires. N''gattamen,\v^zxvi,orv^'via.. Z.
Gunehunga, IV, 33. They tarry. Guneunga, they stay long. Heck.,
Ind. Names, p. 365.
Gunehungtit. IV, 61. They settle. Gunehunga, they stay.
Guneunga. Ill, 12, 20. They tarry. See Gunehunga.
Gunitakan. IV, 62. Long-and-mild. Guneu, long.
Gunokim. IV, 22. Long while fatherly. Guno, snow. Z. Ooch, father.
Gutikuni. Ill, 18. Single night. Gutti, one; nuktogunak, one night.
R. W.
Hackung. I, 2. Above. Hacki, the earth. Z. Hackimk, on or at the
earth. Raf. translates it as hockung, the place above, the sky, heaven.
Camp.
Hakhsinipek. Ill, 17. On hard, stony sea. Achsitt, a. stone; pek, a.
sea. It may mean "stony sea;" but in the connection I think it is
metaphorical " stone-hard," i. e., frozen sea.
Hakik. I, 4. Much land. Hacki, the earth. Z.
Hallemiwis. I, 3. Eternal being. Hallemiwi, eternally. Z.
Hanaholend. V, 24. River loving. Atnhanne, river. H. Ahoala, to
love.
Hattanwulaton. IV, 60. He-has-possession. Hattan, to have ; wula-
ton, to own, to possess.
Huminiend. IV, 25. Corn eater. Pach-hamineu, parched and beaten
corn, R. W., whence our word hominy,
Ikalawit. V, 55. Yonder between. Ikali, thither.
Init'ako. I, 21. Worship snake. Aan, to come; aki, earth. Raf.
derives the suffix from achgook, snake.
Italissipek. IV, 28. Far from the sea. Ikalissi, further, more; pek,
standing water, or sea.
Janotowi. IV, 9. True-maker. IVnutikowi, he keeps watch. Z.
Doubtful.
Jinwis. I, II. Man-being. See note to passage.
238 THE LENAPE AND THEIR LEGENDS.
Kamik. 1,24. Age or foretime. " A^w?]^-^ at the end of words, alludes
to the ground." Baraga, Otch. Die. Gamunk, on the other side of
the water. Z.
Kelik. in, 3. Much. Comp. Kwelik. An intensive prefix.
Kelitgeman. V, 3. Much planting corn. Comp. kelik ; min, corn or
berry.
Kichipek. V, 26. Big sea. Kitschi, great ; pek, a body of still water.
See p. 100.
Kichitamak. ¥,11,36. Big Beaver. Kitschi, ^^dX; tamaque,\)Q'&.st.x.
Kicholen. 111,14. Big bird. Kitchi,^^2Lt; t5chole7is,\i\x&.
Kihillalend. IV, 6. Thou killest some. NihiUan, to kill, k\ thou.
Kimi. I, 21. Secretly. Kimi, privately. Z.
Kiminikwi. IV, 32. Secretly far off. Kimi, privately.
Kinchepend. IV, 55. Sharp he was. Kineti, sharp.
Kipemapekan. V, 47. Big Lake going. Kitschi, great; pek, lake;
aan, to go.
Kitahikan. I, 21. Great ocean. Ill, 17. Of great ocean. Kiiahican,
the sea, ocean. Z.
Kitanitowit. I, 2, 3, 9. God-Creator. See p. 218.
Kitelendam. Ill, 9. Earnestly. To be in earnest. Z.
Kitohatewa. V, 60. Big ships or birds. Kito, great ; haten, he has.
Kitshinaki. IV, 13. Big firland. Kitschi, ^xqzX., ■axi^ shinaki.
Kiwis. I, 17. Thou being. Kitschiwi, truly, verily. Z.
Kiwikhotan. V, 48. Visiting. Kizviken, to visit.
Kolachusien. V, 6. Pretty bluebird. Kola^=zmilit,^rt\.iy. Doubtful.
Kolakwaming. IV, 29. Fine plain at. Wulit, fine, beautiful. The
sense is doubtful.
Kolawil. Beautiful head. IV, 5, 8. IVutit, fine ; zvil, head.
Komelendam. Ill, 11. Having no trouble. To be free from trouble
or care. K.
Kowiyey-tulpaking. Ill, 20. Old turtle land at. Kikey, old. K.
Tulpe, turtle. Doubtful.
Kshakan. I, 7. It blows hard. Ill, 2. It storms. Kschachan, the
wind blows hard. K.
Kshipehelen. II, 16. Water running off. Kschippehellan, the water
flows rapidly, a strong current. Z. Z. also uses higih hilleu, the
waterfalls. Spelling Book, y>. 122.
Kshipehelep. I, 7. It ran off. K schippehelleup, the water ran off.
Zeis. Gram., p. 224.
Ksin. I, 20. Easy. Ksinachpo, he is at leisure.
Kundokanup. IV, 3. Searching when. N'doniken, I seek, or, n'do-
nam. Z.
Kwamipokho. II, 16. Plain and mountain. Klampeecheneu, it is still
or stagnant water. Z.
VOCABULARY. 239
Kwelik. 1,2,4. Much water. 1,7. Deep water. Quenek^kwelek,
long, extended. Z. Compare kelik.
Kwitikwond. IV, 31. Reprover. Quittel, to reprove. Z.
Lakka welendam. Ill, 8. Troubled or afraid. Lachan welendam,
to be troubled in mind. K.
Lamatanitis. V, 44. Lamatan (Huron), friends. See above, p. 16.
Lanewapi. Ill, 19. Eagle manly. Woapalanne, bald eagle. Z.
Langomuwak. V, 60. Friendly they. Langainit winaxti, he looks
friendly. Z.
Langomuwi, V, 54. Friendly he. Langundo, peaceful. Z. From
langan, light, easy.
Langundit. V, 32. Made peace. Langundo, peaceful.
Langundo. V, I. Peaceful. Langundo, peaceful. Z.
Langundowi. IV, 18. Peaceful he. See above.
Lapawin. IV, 40. Whitened. Lappi, again ; pawa, rich.
Lappimahuk. IV, 41. Again there is war. Za//?, again; viachtage-
wak, they are at war. Z.
Lappinup. I, 9. Again when. Mr. Anthony translates this " again he
spoke;" apionen, to SY>&3k.. Zeis.
Lapihaneng. V, 27. Tide water at. Za//z, again; amhatme, flowing
water. H.
Lekhihitin. V, 5. Writer writing. Lekhiket, writer; lekhiken, to
write. K.
Leksahowen. IV, 23. Writing who. Lekhasik, written. K.
Lennowak. I, 11, 18. Men. II, i, 5. Men also. Lenno, man.
Lessin. Ill, 4. To be. Lissin, to be or do so.
Linapi-ma. II, 14. Men there. Lenape, with suffix 7na, there.
Linapioken. IV, i. Men fathers. Qy. "The fathers of the Linapi."
Linkwekinuk. V, 19. Looking well about. Linquechin, to look, be-
hold ; linqtiechinock ! Look here, behold ! Z.
Linnapewi. Ill, i. True manly. Ill, 7. True men. "They are
Lenape."
Linni wulamen. IV, 63. Man of truth. Lenno, man ; 2uulamen.
See p. 104.
Linowi. II, 10. Men. Lenno-wi, he is a man.
Linowimokom. II, 8, 13. Of men grandfather. Lenno, TXizxv; moho-
mus, grandfather.
Lissilma. IV, 5. Be thou there. Lissil, imperative of lissin. Zeis.
Gi-am., p. 118.
Lohxin. II, 9. To move and dwell. Loivin, to pass by. K. Lauch-
sin, to walk, to live. Zeis. Gram., p. 132,
Lokwelend. V, 15. Walker. Lauchsin, to live, to walk.
Lowako. V, 16. North snake. Lowan, winter; aki, land.
240 THE LENAPE AND THEIR LEGENDS.
Lowaniwi. IH, 6, il, i6. Northerlings. Lowan, winter; lowaneu,
north. Z,
Lowanaki. IH, 7. North country. Lowan, winter ; aki, land.
Lowanapi. Ill, 19. Northern manly. Lowan, winter; ape, man; a
nomen gentile.
Lowanipekis. IV, 61. North of the lakes. Lowan,vi'\xA&x\ pek,\zk.^;
or lowan, ape, man; aki, land, "the land of the Northern men."
Lowank warn ink. 111,3. In northerly plain. Zowaw, winter or north ;
weme7ique, as we came from. Z. ; with the locative suffix 7tk.
Lowanuski. IV, 45. Northern foes. Lowan, north or winter.
Lowaponskan. V, 50. North walker. Zijwaw, winter ; north ; /o/wjz'w,
to walk. Z.
Lowashawa. IV, 41 ; V, 59. North and south. Lowan, north ;
skawatio, south.
Lowushkaking. V, 18. North land going. Lo7i> an, north; a>^z, land.
Doubtful.
Luchundi. Ill, 14. They saying. Luehiindi, they say, or, it is said.
Z. Gram., p. 175.
Lumowaki. Ill, 7. White country. Loafnoe, long ago, ancient ; aki,
land.
Lungundowin. II, 3. Peaceful or keeping peace. Langundowi,
peaceful.
Lusasaki. Ill, 10. Burned land. Z/«i-m, to burn ; htsasu,\)ii.Tsx\.. Z.
Machelinik. IV, 58. Many places or towns. Macheli, much. K.
Machigoklos. IV, 38, Big owl. Macheu, great ; goklos, owl.
Machiton. II, 3. Spoiling. Matschihilleu, spoiled. K. Matschiton,
to spoil something, to make mischief. Z. Gt-am., p. 222.
Machitonanep. IV, 17. Much warfare then. Made mischief. See
Ante.
Madawasim. IV, 34. Great meadow. Malta, no, not; assin, siont.
Mahiliniki. V, 46. There was Hilinis. Perhaps " Illini," the Chipe-
ways or Illinois.
Mahongwi. V, 31. There Hong (Mengui) or lickings. Mengwe?
See p. 14.
Mahongwipallat. V, 53. Mengwi was. See last word.
Mahongwichamen. V, 54. Mengwi frightened.
Makatapi. IV, 16. Blacking man. Jlfackit,hz.A, e.\\\; ape, msin.
Makdopannik, V, 4, and Makdupannek, II, 11. They were many.
A/ac/ieli, many.
Makeleyachick. V, 9, Many going. See above.
Makelohok. IV, 48. They are many. See above.
Makeliming. V, 6. Much fruits at. Macheletnuwi, honorable, pre-
cious. K. Ox macheli, nxMch; »««, fruits.
VOCABULARY. 241
Makelining. V, 8. Much river at. Machelensin, to be proud or high-
minded. K. Or, macheli, much or many; amhanne, rivers, "the
place of many streams."
Makelima. IV, 56. Much there is. Macheli, much or many.
Makelinik. V, 7. Many towns. Macheli, mzxiy; wik, \io\x5ts.
Makeliwulit. V, 38. Much good done. Macheli, much ; wulit, good.
Makelomush. V, 41. Much honored. Machelemtixit, he that is hon-
ored. Z.
Makhiawip. V, 27. Red arrow. Machke, red.
Makimani. I, 14. Bad spirit. Machi manito, the bad manito.
Makonowiki. V, 46. There was Konowis. Qy. Achgunnan, he is
clothed. Z. Mach, = red ; mecaneu, dog.
Makowini. I, 14; II, l. Bad beings. Mach, from machtit, bad;
owini, q. v.
Makpalliton. V, 15. Much warfare. Macheli, much, and palli-
ton, q. V.
Maktapan. I, 23. Bad weather. Machtapan, stormy weather. K.
Maktaton. I, 22. Unhappiness. Machtatemantoagan, unhappiness. K.
Mangipitak. IV, 22, Big teeth. Atnangi, big, great ; wipit, his
teeth.
Mani. I, 8. Made. Maniton, to make.
Manito. I, 9, 10. He made. II, 12. Spirit. See notes.
Manitoak. I, 9, 17. The spirits or makers.
Manup. IV, i. There were then. Doubtful. Comp. anup.
Mapawaki. V, 22. There is rich land. Fawa, rich; aki, land.
Doubtful.
Mashawoniki. V, 46. There was Shawonis. Meshe, great, in comp.
Mashkipokhing. IV, 7. Bear hills at. Machk, bear; but probably
from maskiek. Chip, mashkig, swamp or marsh, and pachhink, the
division or valley between the mountains.
Maskaboush. II, 8. Strong hare. Maskan and wabos, hare. See
ante, p. 130.
Maskan. II, i, 2, 5, 16. Powerful or dire. Mechek, great, large;
mangain, Nant. mashka. Chip, strong. Maskane, strong, rapid.
Heck., Ind. Names, p. 355.
Maskanako, II, i, 2, 5. Strong snake. Maskan, large or strong;
achgook, snake.
Maskansisil. IV, 37. Strong buffalo. Maskan, and sisil.
Maskansini. IV, 43. Strong stone. Maskan, and assin, a stone.
Maskekitong. V, 28. Strong falls at (Trenton). Maskan, z.x\A kiihanne,
main stream. See Heck., Ind. Names, p. 355, where this word is
given and analyzed.
Matemik. IV, 20. Builder of towns. Malla, not; meguik, hlood. Z.
Matta. II, 3. Not. Malta, no, not.
242 THE LENAPE AND THEIR LEGENDS.
Mattakohaki. V, 22. Without snake land. Matta, not; achgook,
snake ; aki, land.
Mattalogas. I, 22. Wickedness. Machtit, bad, evil; niattalogaso-
wagon, a sinful act. Zeis. Grarn., p. 103.
Mattapewi. II, 4. Less man. Mattapeti, he is not at home. Z.
Matemenend. IV, 36. There 07- now Tamenend. .
Mawulitenal. V, 22. There is good thing. Wtilit, good.
Mayoksuwi. IV, 53. Of one mind. Alazvat, one, only one. K.
Mboagan. I, 23. Death. APboagan, death. Z.
Mekenikink. I, 21. On earth. Much, prefix indicating evil or mis-
fortune, from tnachtit.
Mekwazoan. II, 4. Fighting. Mechtagan, to fight. K.
Menak. I, 8. Islands. Alenafey, an island.
Menalting. IV, 4,42. In assembly met. yJ/«;ar////«, to drink together. K.
Alendltink, the place vi^here we drank. H. Ind. Names, p. 371.
Menapit. 11,8. At that island. Menaiey,\s\^xiA; epil,sX.
Meshautang. Ill, 3. Game. Mechtit, much; achtu, deer. Z. In
the N. J. dialect, deer is aatu ; hence the meaning is "many deer."
Messisuwi. IV, 44. Whole he. Metschi-schawi, very, ready. Z.
Metzipannek. II, 11. They did eat. Milzopantii k, ih&y have eaten,
Zeis. Gram., p. 124.
Michihaki. IV, 3. Big land. Mechti, much ; aki, land.
Michimini. IV, 34. Much corn. Mechtil, -awiohy, wm, edible fruit.
Milap. 1,12,13. He gave him. Alil or miliin, io g^\t. The terminal
/ marks the preterit.
Minigeman. IV, 25. Corn planting. Mm, edible fruit ; for corn, see
p. 48.
Minihaking. IV, 24. Corn land at. Min, edible fruit ; aki, land.
Minsimini. V, 52. Wolf tribe. See p. 36.
Mitzi. I, 19. Food. Mitzin, to eat.
Mokol. 11,12. Boat. Amochol, :i.ho^i. Zeis. Gram.,Y>- loi.
Mokolakolin. V, 17. In boats he snaking. See above. Aki, land.
Mokom. V, 17. Grandfather. Mtickomsena, our grandfather. Z.
Mokolmokom. V, 17. Boats grandfather. Amochol, boat; mtichom,
ancestor.
Moshakwat. I, 7. It clears up. Aloschhakquat, clear weather. K.
Mukum. I, II, Ancestor, .i^/wc/^ow^j, grandfather. K.
Nahiwi. II, 10. Above water or afloat. Nahiwi, down the water,
down stream. K.
Nakhagattamen. V, 52. 3 desiring. Nacha, three ; gatta?nen, to
wish.
Nakkalisin. V, 52. 3 to be. A^acha, ih.xeQ; /issin, to he or do so.
VOCABULARY. 243
Nakopowa. Ill, 8. The snake priest. Patua, priest. See above, p.
70. The prefix doubtful.
Nakowa. II, 6. Black snake. Nachoak, three persons. Z.
Nakowak. I, 14. Black snakes. Nachohaneu, he is alone. Z. Suk-
achgook, black snake. Z. Doubtful.
Nallahemen. Ill, 13. Navigating. Nallahemen, to boat up the
stream. K.
Nallimetzin. IV, 29. At last to eat. Nail, that, at last; mitzin, to eat.
Namenep. I, 20. Pleased. Namen, to know, understand.
Namesaki. IV, 14. Fish land; Namaes,^%\^; «/^2, land.
Namesik. I, 13. Fishes. Namessall, fishes. Zeis. Gram., p. loi.
Namesuagipek. Ill, 12. Fish resort sea. Namaes,'i\^\ pek,\'sk^.
Nanaboush. II, 8, 13. Nana-hare. See p. 130.
Nantine. I, 19. The fairies. Naten, to fetch. Z.
Nantinewak. I, 18. Fairies also. PI. form from naten, to fetch.
Nekama. IV, 9, 10, 19. Him. Him, them.
Nekohatami. IV, 35. Alone the first. Netatni, the first.
Nemassipi. IV, 49. Fish river. Namaes,^^; sipi,xvitx.
Nenachihat. V, 58. Watcher. Nenachgistawachtin, to listen to one
another, to hear one. K. Hence hearer.
Nentegowi. V, 16. The Nentegos. Nentego is the proper name of
the Nanticokes, who inhabited the eastern shore of Maryland. See
p. 22.
Netamaki. I, 24. First land. Netatni, first ; aki, land.
Netami. I, 12, 18, 19. The first. Netami, the first. Z. Gram., p. 108.
Nguttichin. Ill, 16. All agreed. ' Nguttitehen, to be of one heart
and mind. Z.
Nigoha. I, 18. Mother. Ngahomes, my mother. See Zeis. Gram.,
p. 100.
Nihantowit. 11,4. Dead keeper. ' Nihillowei, mmdertx {nihillajiowet).
See p. 102.
Nihillanep. IV, 43. He killed. See p. 102.
Nihillapewin. Ill, 11. Being free. Nihillapewi, free. Z. See
p. loi.
Nihillen. Ill, 15. To kill or annihilate. Nihilla, I kill. Z. See
p. lOI.
Nijini. I, 10, 19; 11,2. The Jins. iV?^, these, those. K. Nigani,
the first, the foremost. Z. See notes.
Nillawi. HI, 18. By night or in the dark. Nipahwi, by night. Z.
Nipahum. I, 5. Moon. Nipahump, moon, Min.
Nishawi. II, 3. Both. Nischa, two.
Nitaton, IV, 11. To be able. To know how to do it. Z.
Nitatonep. IV, 43. He was able. See above. Preterit.
244 THE LENAPE AND THEIR LEGENDS.
Nitisak. I, i6. Friends. Nitis, confidential friend. (Heck, p. 438.)
Nitilowan. IV, 54. Friends of north. Nitis, and lowan, north.
Nolandowak. IV, 49. Lazy they. Nolhand, lazy. K.
Nolemiwi. I, 3. Invisible. Invisible. Z.
Nungihillan. Ill, 10. By trembling. Nungihillan, to tremble. K.
Nungiwi. IV, 64. Trembling he. See above.
Okwewi. I, 18. Wives. Ochquewak, women. Z.
Okwisapi. Ill, 19. With vi^ives or women of man. OcJiqiie, woman;
ape, man.
Oligonunk, IV, 29. Hollow mountain over. Wahlo, a cavern or a
hollow between hills. Oley, in Berks county. Pa., the name of a
Moravian settlement, is from this root.
Olini. Ill, 18. The men or people. From root ni, p. loi.
Olumapi. IV, 23. Bundler of written sticks. See p. 161.
Onowutok. V, 12. Prophet. Owoatan, to know. K.
Opannek. Ill, 16. They went. From aan, to go, and perhaps with
prefix wab or op, east.
Opekasit. IV, 47. Easterly looking. Waopink or opilnk, opossum.
From the root wab, white. See p. 43.
Opeleken. I, 8. It looks bright. Root ivab or op. See last word.
Otaliwako. V, 43. There snake or Otalis (Cherokis).
Otaliwi. V, 56. Cherokees of Mts.
Ouken. Ill, 12. Fathers. Ochwall, his father. Zeis. Gram., p. loo.
Owagan, I, 22, or Owagon, I, 7. Deeds, action. A verbal suffix. See
p. lOI.
Owak. I, 4. Much air or clouds. An error for woak, and. Comp. Zeis.
Spelling Book, p. 122.
Owanaku. I, 2. Foggy. Awonn. Z. Atian, N. J., fog.
Owini. I, 12. First beings. I, 16; 11,5,9. Beings. Rafinesque says
of this word, that it "may be analyzed o-wi-ni, 'such they men' or
beings." It would seem to be a form of the substantive verb termi-
nation wi.
Owinkwak. I, 10. First beings also. Owini, and wak, and.
Paganchihilla. IV, 59. Great fulfiller. Pachgikillan, to break, break
asunder. K.
Pakimitzin. V, 49. Cranberry eating. Pakihin, cranberries; mitzin,
to eat.
Pallalogas. I, 22. Crime. Pallalogosawagan, crime, evil deed. Zeis.
Gram., p. 103.
Palliaal. Ill, 9. Go away. The same. Zeis. Gram., p. 243. An
imperative ; but not so used in the text.
VOCABULARY. 245
Pallihilla. IV, 56. Spoil and killing. From pallilissin, to do wrong.
Zeis. Gram., p. 243.
Palliton. II, 3. Fighting. II, 5. To destroy or spoil. II, 7. Much
spoiling or destroying. Palliton, to do ill, to spoil. Zeis. Gram.,
p. 222.
Pallitonep, IV, 44, 46. He war made. It is the imperfect Qi palliton,
to despoil, fight.
Pallitonepit. IV, 47. At the warfare. Preterit of the above.
Palliwi. II, 16. Elsewhere. Ibid. Z.
Palpal. II, 12. Come, come. Palite, when he comes. Z.
Paniton. II, 15. Let it be. Paliton, to spoil, injure. Z.
Pataman. II, 15. Praying. Pataman, to pray. K.
Pawanami. V, 14. Rich water turtle. Pawalessin, to be rich.
Pawasinep. Ill, 13. Rich was. Pawa, rich.
Payat. I, 23. Coming. Paan, to come. Conjugated in Zeis. Gram.,
p. 148. Payat, he who comes or is coming. From the root an, to
move. Cf. Aan.
Payat-chik. I, 22. Coming them. See above.
Payaking, III, 20. Coming at. See above.
Payat payat. II, 12. Coming, coming. See above.
Pechimin. Ill, 10. Thus escaping. Pack-, to separate, divide, to split
asunder.
Pehella. II, 7. Much water rushing. II, 10. Flood. See Kschippe-
hellen.
Peklinkwekin. V, 59. Sea looking. Pek, still water, lake, sea.
Pekochilowan. V, 23. Near north. Lowan, north.
Pemaholend. IV, 20. Constantly beloved. Ahoala, to love.
Pemapaki. IV, 14. Lake land. Apparently for memippekink, at the
lake.
Pematalli. V, 17. Constant those. 7a//z, there.
Penauwelendamep. II, 5. Resolved. Penauwelendam, to consider
about something. Z.
Penkwihilen. II, 16. It is drying. Penqtdhillen, dried. K.
Pepomahemen. V, 8. Navigator up. Doubtful.
Petonep. II, 6. He brought. Peton, to bring. Z.
Peyachik, III, 4. Comers. See Payat.
Pikihil. Ill, 10. Is torn. Pikihillen, torn, rent in pieces. K.
Pilwhalin. IV, 21. Holy goer. Pilhik, clean, pure.
Pimikhasuwi. IV, 57. Stirring about he.
Piskwilowan. V, 31. Against north. Tipis qui, a.ga.mst. Z. Lowan,
north.
Pitenumen. V, 39. Mistaken. Pitenummen, to make a mistake. Z.
246 THE LENAPE AND THEIR LEGENDS.
Pohoka. II, 7. Much go to hills. Pokawachne, creek between two
hills. The word does not refer to hills, but to the division, cleft or
valley between hills.
Pokhapokhapek. Ill, 12. Gaping sea. Pocqiieu, a muscle, clam. Z.
An important article of food to the natives ; pek, a lake or sea.
Pokhakhopak. Ill, 17. At gap snake sea. See above.
Pokwihil. Ill, 4. Divided or broken. Ill, 10. Is broken. Poqui-
hilleii or poqidecheu, broken. K. The root is pack, to split, divide.
Pomisinep. IV, 52. Went or passed. Pomsin, to walk. K.
Pommixin. II, 9, 10. Creeping. Po7nmisgeii, to begin to walk ; pom-
^mixin, to creep. K.
Ponskan. Ill, 18. Much walking. Pommatichsin, to walk.
Powa. HI, 4. Rich, for Pawa, rich, etc. See p. 70. See words under
pawa.
Powako. I, 21. Priest snake. See above.
Powatanep. IV, 39. Pontiff was. See above.
Powatapi. Ill, 19. Priest manly. See above.
Psakwiken. HI, i. Close together. Psakquiec/ien,c\osziogti\\er. K.
Pungelika. V, 31. Lynx well like (Eries). Pongus, sand fly. K.
Doubtful.
Pungusak. I, 15. Gnats. Pojigiis, sand fly, K.
Sakelendam, IV, 47. Being sad. Sakqiielendam, to be sad. K.
Sakima. IV, 5. King. See p. 46.
Sakimachik. IV, 26. See above.
Sakimak. IV, 17. Kings. See above.
Sakimakichwon. V, t,t,. With this great king. See above.
Sakimalanop. IV, 33. King was made. See above.
Sakimanep. IV, 8, 9, 15, 18. King was. See above. Preterite form.
Saskwihanang. V, 24. Susquehanah (branchy R.) at. See p. 14.
Sayewis. I, 3. First being. Schawi, immediately, directly. Z.
Shabigaki. IV, 13. Shore land. This seems a more correct form than
Heckewelder's scheyichbi. See p. 40.
Shak. I, 14. But. ScJiiik, but.
Shakagapewi. IV, 64. Just and upright he. Schachachgapewi, he is
honest, righteous. K.
Shakagapip. IV, 19. A just man he was. Schachach, straight; here
used in a metaphorical sense for just.
Shawaniwaen. IV, 12, 24. South he goes. Shawano, south
Shawanaki. IV, 13. South land. Shawano, south; «/^z, land. Zeis.
gives schawenneu for south.
Shawanaking. V, 10. South land at. See above.
VOCABULARY. 247
Shawanapi. Ill, 19. Southern manly. Shawano, and ape, man.
Shawaniluen. IV, 10. South he saying. Shawano, and luen, to say.
Shawaniwak. IV, 59. South they go. Shawano, and ewak.
Shawanipalat. V, 42. South warrior. Shawano, and itapalat.
Shavvanipekis. IV, 60. South of the lakes. Shawano, and/^/§, lake.
Shawaniwi. Ill, 6. Southerlings. Shawano, with suffix wi.
Shawanowi. V, 10. The Shawani. See above.
Shawapama. IV, 17. South and east there. Shawano, wapan, east,
and ma, there.
Shawelendamep. II, 2. Become troubled, Acqiiiwelendam, tq dis-
quiet. Z. With intensive prefix ksch.
Shawoken. Ill, 10. So far going, ^ir/z^wai, weak?
Shayabinitis. V, 57. Shore friend. See next words. Nitis, friend.
Shayabian. V, 37. Shore (or Jersey) going. Schejek, a string of
wampum. Z.
Sheyabing. V, 51. At New Jersey <7r shore. ^ir/^^jK/i^/^^?, Indian name
of New Jersey. (Heck., p. 51.) See p. 40.
Shinaking. Ill, 20; IV, i, 5. At fir-land. Chip, jin-goh, spruce fir.
Bar. Schind, spruce. Z. Aki, land ; nk, locative termination, " the
place of spruce firs."
Shingalan. II, 2. Hating. Schingalan, to hate somebody. K.
Shingalusit. II, 2; V, 56. Foe, foes. Schingalusit, enemy, adver-
sary. K.
Shiwapi. IV, 27. Salt man. Schwewak,%'zS.\.xix^z.t; sthey, salt.
Showihilla. IV, 7. Weak. Schawek, weak.
Shukand. I, 20. But then. Schukund, only, but then.
Sill. HI, 3. Cattle. Sisili, a buffalo. See note to verse.
Sin. HI, 4. To be. Lissin, to be or do so.
Sinako. V, 16. Strong snake. ^j«;z, stone; «/^z', land.
Sipakgamen. IV, 55. River over against. Sipi, river. See Agamunk.
Sisilaki. IV, 14. Cattle land. Sisilianiuus, a buffalo, N. J.
Sisilaking. IV, 29. Cattle land at. Sisili, buffalo ; aki, land.
Sittamaganat. V, 2. Path leader. Pipe-bearer. See note to IV, 2.
Sitwahikho. II, 16. Path of cave. Tschitqui, silent; tschitquihille-
wak, they are silent. Z.
Slangelendam. IV, 31. Disliking. Skattelendam, to loathe, to hate.
Sohalawak. I, 4, 5, 6, 14, 15 ; IV, 23. He causes them. See note,
Sohalgol. IV, 25. He causes it. See last word.
Taquachi. IV, 24. Shiverer with cold. Tachquatten, frozen. K.
Takauwesit. HI, 5. The best. Tach, together, to tie, etc. Hence
united, harmonious.
Talamatan. IV, 54, 61, 63, 64. Hurons. See p. 16.
248 THE LENAPE AND THEIR LEGENDS.
Talamatanitis. IV, 6i. Huron friends. See Lamatanitis.
Talegachukang. V, 19. Allegheny Mts. going. Doubtful.
Talegaking, V, i. Talega land at. See p. 230.
Taleganah. V, 14. Talega R, at. See p. 230.
Talegawik. IV, 56. Talega they. See p. 230.
Talegawil. IV, 52. Talega head or emperor. See p. 230. IVil,
head.
Talegawunkik. V, 45. Talegas west visitor. See p. 230. Wunken,
west ; kiwiken, to visit.
Talligewi. IV, 50. Talegas or there found. See p. 229.
Tamaganat. IV, 55. Leader. Gelekmend ^=ihQ leader. Heck. Ind.
Names, p. 392. See note to IV, 2.
Tamaganena. V, 2. Chieftain such or Beaver leader. Pipe-bearer.
See note to IV, 2.
Tamakwapi. Ill, 19. Beaver manly. Tamaqiie. Camp. Ktemaque.
Zeis. A beaver. Mohegan, amuchke, Schmick.
Tamakwi. IV, 12. Beaver he. See last word.
Tamenend, IV, 35 ; Tamanend, V, 32. Affable (beaver like). Teme-
nend, affable. Heck.
Tankawun. V, 9. Little cloud. Tangelensuwi, modest, humble ;
tangitti, small.
Tapitawi. II, 14. Altogether. Tachguiwi, together. Z.
Tashawinso. V, 51. At leisure gatherer.
Tasukamend. IV, 19. Never black or bad. Ta, not, suckeu, black.
Z.
Tatalli. II, 10. Which way or shall there. 7a/a//z, whitherwards. K.
Tawanitip. V, 49. Ottawas made friends; nitis, friend.
Tellen. IV, 17. Ten.
Tellenchen kittapakki. Ill, 18. 10,000.
Tenche kentit. IV, 58. Opening path. Tenk, titit, little. K.
Doubtful.
Tendki. 111,8. Being there. Tindey,^XQ. Z. Tenden, Min.; yawa-
gan tendki, the cabin-fires.
Tenk wonwi. IV, 27, 30. Dry-he. Teng- or tenk- ;= little. K.
Thupin. Ill, 2. It is cold. Teti, it is cold. K.
Tihill. Ill, 3. Coolness. Tillihan, it is cool. K.
Topan. Ill, 2. It freezes. Tepan, white frost.
Topanpek. Ill, 16. Frozen sea. 7>/a«, and ; /^^, lake.
Towakon. IV, 46. Towako. V, 16. Father snake. Tawa and aki,
the Ottawas or Twightees. See note to V, 16.
Tsehepicken. IV, 49. Separated. Tschetschpiechen, to separate. K.
Tulagishatten. II, 9. At Tula he is ready. Tulpe, turtle ; gischatten,
it is ready, done, finished.
VOCABULARY. 249
Tulamokom. II, 13. A turtle's grandfather. Tidpe, turtle. See
Mokom.
Tulapewi. II, 14. Turtle there. Tidpe, a water turtle. K.
Tulapewini. Ill, i. Turtle being. See above.
Tulapima. II, 14. Turtle there. Tulpe, and ma, there.
Tulapin. II, 10. Turtle-back. Tulpe, turtle.
Tulapit. II, 8. At Tula or turtle land. Tulpe, and epit, q. v.
Tulapiwi. Ill, 7. The turtling. Tulpe, and suffix wi.
Tulpenaki. Ill, 7. Turtle country. Tulpe, and aki, land.
Tulpewi. 11,15. Turtle he. See above. Tulapezvi.
Tulpewik. I, 13. Turtles. See above.
Tumaskan. IV, 42. Wolf strong. Temnieu, vcolf, Z.
Tumewand. V, 29. The wolfers (mohican). Temineu, wolf, anit =
the wolf god, or magician.
Tumewapi. Ill, 19. Wolf manly. Temmeu, and ape man; a nomen
Uchewak. I, 15. Flies. Utschewak, flies. Z.
Unamini. V, 52. Turtle tribe. See p. 36.
Unchihillen. V, 39. Coming from somewhere. Untschihilleu it come s
from somewhere rapidly, to flow out.
Wagan. II, 16. Action. See Oiuagan.
Wak. I, 2. And. Id.
Wakaholend. IV, ■^1. Loving, beloved. Ahoalan, to love. Woaka-
holend. Heck. Ind. Names, p. 395.
Wakon. I, 21. Snake god. Wachunk, high (Min.) Perhaps a form
of akiuk, earthward.
Wallama. IV, 40. Painted. See p. 161.
Wallamolumin. V, 5. Painted-booking. See p. 161.
Wangomend. V, 55. Saluted. Id. Heck. Ind. Names, p. 395.
Wapachikis. V. 57. White crab. Woapeu, white. Z. The root
wab, wap, or op, white, light, the east, etc., occurs in numerous
words.
Wapagumoshki. V, 44. White otter. See above.
Wapagishik. IV, 48. East sun or sunrise. Wap, anA gischuch.
Wapagokhos. IV, 8. White owl. Wap, and gokhos, owl. Z.
Wapahacki. V, 37. White body. Wap, and hackey, body.
Wapahoning. V, 11. White Lick at. Wap.and. mahoning. Z. A
the deer lick.
Wapakisinep. V, 21. East land was. Wap, and aki, land, with pre-
terit suffix.
Wapalaneng. V, 2. White river at. Wap, and amhannink at the
river.
Q
250 THE LENAPE AND THEIR LEGENDS.
Wapala wikwan. V, 20. East settling place. Wap, and zuikwam,
house.
Wapallanewa. IV, 2. White eagle. Woaplanne, the bald eagle. Z.
Wapallendi. IV, 52. East some, /f^^;/, east; aHende, somQ.
Wapanaki. Ill, 18. Eastern land. Wap, east; aki, land.
W^apanapi. Ill, 19. Eastern manly. Wap, east or white; ape, ma.n.
Wapaneken. IV, 48. East going together. Wap, east ; see Eken.
Wapanen. Ill, 9. Easterly. Wap, east.
Wapanand. V, 29. The easters. Wap, east.
Wapanichan. IV, 32. East moving. Wap, east.
Wapaniwaen. IV, 12, 28. East he goes. Wap, east; aan, to go.
Wapaniwi. 111,6, 16. Easterlings. Wap, east; wi, substantive verb
suffix.
Wapashum. V, 45. White big horn. Wap, white ; wschummo, horn.
Z.
Wapasinep. Ill, 13. East was t;;- bright. JFap, east; preterit termi-
nation.
Wapawaki. IV, 51. East rich land.
Wapawullaton. IV, 50. East possessing. Wap, east ; zvitllaton, to
possess.
Wapayachik. V, 59. White or east coming. Wap, east ; payat, q. v.
Wapekunchi, V, 40. East sea from. Wap, east ; doubtful.
Wapkicholan. IV, 38. White crane or big bird. Wap, white; tscho-
len, bird.
Waplanowa. Ill, 12. White eagle. Woaplanne, a bald eagle. Z.
Waplowaan. V, 29. East, north, do go. Wap, east; lowan, north,
aan, to go.
^A^apsipayat. V, 40. Whites coming. Wap, white ; payat, q. v.
Waptalegawing. V, 20. East of Talega at. Wap. east ; talega, q. v.
Waptipatit. IV, 41. White chicken. Wap, white; tipatit, chicken.
Waptumewi. Ill, 12. White wolf. /r«/, white ; /^ww^m, wolf .
Wapushuwi. V, 3. White lynx he. Wap, white.
Wasiotowi. V. 56. Wasioto. Doubtful.
W'delsinewap. I, 16. Were there. Preterit of lissin, to be so.
Wekwochella. IV, 30. Much fatigued. Wiqiiehilla, to be tired. Z.
Wellaki. IV, 3. Fine land. Wulit, fine ; aki, land.
Wemaken. Ill, 15. All snaking. Wemi,3.\\; aki, land, earth; the
whole land.
Wematan. Ill, 14. All let us go. Wemi, and ata7n, q. v.
Wemelowichik. V, 26. All hunters. Wemi, all; elauwitschik,
hunters.
Wemi. I, 7, 6, 16, 20. All. Id.
VOCABULARY. 251
Wemiako. Ill, 8. All the snakes. IFemi, all ; achgook, snake ; or,
aki, land.
Wemiamik. V, 48. All children (Miamis). Doubtful.
Wemichemap. II, 12. All helped. Wemi, all; mitschemuk, he
helps me. Z.
Wemiguma. I, i. Wemi, all; gutna, sea water. See note to passage.
Wemiluen. Ill, 15. All saying. Wemi, all; hien, to say.
Wemimokom. II, 13. Of all grandfather. Wemi, and tnokom, q. v.
Wemilowi. IV, 53. All say. Wemi, all ; bieti, to say.
Weminitis. IV, 35. All being friends. V, 33. All friendly. Wemi,
all; nitis, friends.
Wemipalliton. IV, 43. To war on all. Wemi, and palliton, q. v.
Wemima, IV, 2. All there. Wemi, all; ;««, there.
Wemilat. IV, 58. All given to him. Wemi, and miltin, q. v.
Wemilo. IV, 5. All say to him. Wemi, and lueti, to say.
Weminilluk. IV, 15. All warred. Wemi, and nihillan, q. v.
Weminitik. V, 48. All friends or allies. Wetiii, and niiis.
Weminungwi. V, 31. All trembling. Wemi and mtngihillan, to
tremble.
Wemi owenluen. Ill, 8. To all saying. Wemi, and hien, to say.
Wemi tackwicken. V, 33. All united. Tachquizui, together.
Wemiten. Ill, 11. All go out. IV, 54. To go all united. Wemi-
ten (infin), to go all forth or abroad. Z, Gr. 244.
Wemoltin. II, 10. All go forth. Ill, 9, 18. They go forth. They
are all going forth. Z. Gr. p. 244.
Wemopannek. Ill, 17. All went. Wemi, with past preterit suffix.
Wenchikit. V, 52. Offspring. Wentschiken, to descend, to grow
out of. Z.
Wetamalowi, IV, 33. The wise they. Wewoatamamine, wise man. Z.
Wewoattan. IV, 42. To be wise or by wise. Woaton, to know. Z.
Wich. I, 7. With. Witschi, with.
Wichemap. II, 12. Helped. Wiischeman, to help somebody.
Wihillan. I, 23. Destroying or distemper. Nihillan, to destroy.
Wihlamok. Ill, 14. Head beaver. Wil, head ; atnuchke, beaver.
Moh.
Wikhichik. Ill, 4. Tillers. Wikhetschik, cultivators of the earth. Z,
Wiki. II, 4. With. Witschi, with.
Wikwan. V, 20. Wikwam, house.
Wilawapi. 111,19. Rich manly. Wil, hea.d; aJ>e,Taa.n.
Winakicking. V, 25, 27. Sassafras land at or Penna. Winak, sassa-
fras. Z.
Winakununda. V, 36. Sassafras tarry. Winak, sassafras, guneunga,
q. V.
252 THE LENAPE AND THEIR LEGENDS.
Winelowich. V, i8. Snow hunter. Wineu, snow; elauwzVsc/i, hunter.
Wineu. Ill, 2. It snows. Wineu, it snows.
Wingelendam. IV, 60. Wingelendam, to approve, to like. Z.
Wingenund. IV, 39. Mindful.
Wingi. I, 20. Willingly. Wingi, fain, gladly, willing.
Winiaken. Ill, 11. At the land of snow. Wineu, it snows; aki,
land.
Winimokom. II, 13. Of beings grandfather. Ozvini and Mokom, q. v.
Wisawana. IV, 34. Yellow River. Wisaweu, yellow; amhanne,
river.
Wishanem. II, 15. * Frightened. Wischaleu, he is frightened. Z.
Wishi. I, 17. Good. Probably for mesitche ^^QVx"^, ?nitcha, etc., great,
Witchen. 111,15. Going with. Witen, to go viiih.. K.
Wittank. IV, 34. Town. Witen, to go or dwell with.
Wittanktalli. Ill, l. Dwelling of Talli. Witen, to go with. Z.
talli, there. Z.
Wiwunch. I, 24. Very long. Wiwuntschi, before now, of old. K.
Wokenapi. IV, 1 1. Fathers men. ^Foa/^/a/// repeatedly, again. K.
Wokgetaki. I, i. Wokget, on the top; aki, land. Wochgitschi,
above, on top ; aki, land, earth.
Woliwikgun. Ill, i. Cane house. Walak, hole; walkeiijhe'vi dig-
ging a hole. Z.
Wolomenap. V, 28. Hollow men. Wakkillejnato, wide, far. K.
Won. I, 24. This. Woft, this, this one. K.
Wonwihil. V, 40, 59. At this time. Won, this, wil, head.
Wsamimaskan, IV, 57. Too much strong. Maskan, great.
W'shakuppek. Ill, 17. Smooth deep water. Wschacheu, it is slip-
pery, smooth, glossy ; pek, lake, sea.
Wtakan. Ill, 3. Mild. Wtakeu, soft, tender. Z.
W'tamaganat. IV, 37. And chieftain. The smoker or pipe bearer.
See note to IV, 2.
Wtenk. I, II. After. Ibid.
Wulakeningus. V, 42. Well praised. Wulakenimgussin, to be
praised. K.
Wulamo. II, I; IV, i; V, i. Long ago. Wulafnoe,\Qn'gz.go.
Wulaton. 111,3; IV, II. To possess.
Wulliton. Ill, i6. Wulaton, to save, to put up. K. Wuliton, to
make well. K.
Wulatenamen. V, 41. To be happy. Ibid.
Wulelemil. Ill, 17. Wonderful. Wulelemi, wonderful.
Wuliton. II, 15. To make well, to do well. Z. Gr. p. 222.
Wulitowin. IV, 20. Good who (did). See last word.
VOCABULARY. 253
Wulitshinik. V, 4. Good stony <7r well, hardy. IViiHf, good; assin,
' stone.
Wulitpallat, V, 30. Good warrior. Wulii, good ; itopallat, warrior.
Wunand. I, 17. A good god. Root Wim. See p. 104.
Wundanuksin. IV, 32. Being angry. Wiindanuxin, to be angry at
or for. K.
Wunkenahep. V, 12. West he went. WHndcJiene2i,\i\s, vi&zt.
Wunkenapi. Ill, 20. Western man. Wimdc/ien,vfQS\.; ape,mz.n.
Wunkeniwi. Ill, 6. Westerlings. See above.
Wunkiwikwotank. V, 13. West he visited. See above. Khuichen,
to visit.
Wunpakitonis. V, 13. West abandoned. Pakiton, to throw away.
Wunshawononis. V, 13. West southerners. Shawano, south.
Yagawran. Ill, 8. (In the) huts. Ibid.
Yagawanend. IV, 50. Hut maker. See last word.
Yuch. I, 6. Well. Yuh. H. Yuch. K. Yttk, these. K.
Yukepechi. IV, i. Till there. Yukepetschi, till now, hitherto. K.
Yuknohokluen. IV, 48. Let us go saying. Doubtful.
Yulik. I, 6. These. Yukik, these. K.
Yutali. I, 2, 22. There. Jutalli, just here. K.
APPENDIX.
AGOZHAGAUTA. {page 1 4, Note.)
With reference to this word I have been favored with the opin-
ions of Gen. Clark, Mr. Horatio Hale, and the Rev. J. A. Cuoq,
all able Iroquois scholars.
Gen. Clark and Mr. Hale believe that it is a dialectic or corrupt
form for agotsaganha, which is a derivature from atsaganneji
(Bruyas, Radices Verborum Iroquceorum, p. 42). This verbal
means, in one conjugation, " to speak a foreign language," and
in another, " to be of a different language, to be a foreigner." The
prefix ago or ako is an indefinite pronoun, having the same form
in both singular and plural, and is used with national or tribal
appellations, as va. akononsionni, "People of the Long House,"
the general name of the Five Nations. Gen. Clark notes that the
term agotsaganens, or agotsaganes, was the term applied by the
Iroquois to the Mohegans, = " People who speak a foreign tongue."
(Jogues, Novum Belgium (1646), and Pa. Colonial Records, vol.
vi, p. 183.)
The Rev. Mr. Cuoq believes that the proper form is akotsa-
kannha, which in his alphabet is the same as agotsaganha, but he
limits its meaning to " on est Abnaquis," from aktsakann, "etre
Abnaquis." (See his Lexique de la Langue Iroquoise, pp. i, 155.)
The general name applied by the Iroquois to the Algonkins he
gives as Ratirontaks, from karonta, tree, and ikeks, to eat, " Tree-
eaters" {Lexique, p. 88) ; probably they were so called from their
love of the product of the sugar maple.
DIALECT OF THE NEW JERSEY LENAPE. {p. 46)
An interesting specimen of the South Jersey dialect of the
Lenape is preserved in the office of the Secretary of State, Tren-
ton, N. J. It is a hst of 237 words and phrases obtained in 1684,
at Salem, N. J. It was published in the American Historical
Record, vol. I, pp. 308-311, 1872. The orthography is EngHsh,
and it is evidently the same trader's jargon which Gabriel Thomas
gives. (See p. 76.) The r is frequent ; man is renus leno; devil
is manitto; God is Jiockimg tappin (literally, "he who is above").
There are several typographical errors in the printed vocabulary.
255
256 APPENDIX.
REV. ADAM GRUBE. (/. 84.)
His full name was Bernhard Adam Grube. Between 1760-63
he was missionary in charge of the Moravian mission at Wech-
quetank, Monroe County, Pa., and there translated into Delaware,
with the aid of a native named Anton, a " Harmony of the
Gospels," and prepared an " Essay of a Delaware Hymn Book."
Both these were printed by J. Brandmiiller, at Friedensthal, Pa.,
and issued in 1763 ; but no copy of either is known to exist.
EASTERN ORIGIN OF THE ALGONKINS. {pp. 12 and I4S-)
Quite recently M. Emile Petitot, in an article entitled, De la
pretendue Origi7ie Orientale des Algonquins'' {Bulletin de la
Societe d' Anthropologie , 1884, p. 248), has attacked the theory
that the Algonkin migrations were from the northeasterly por-
tions of the American continent, toward the west and south. His
arguments are based on two Cree legends which he relates, one of
which is certainly and the other probably of modern date, as the
incidents show ; and on his criticism of the derivation of the name
" Abnaki." Of this he says : " IVaiang signifie plutot detroit que
orient ; et quant au mot as^/y ou ahkiy, il vent dire terre, et non
pas peieple."
Now, no one ever claimed that abnaki meant eastern people.
The Abbe Maurault translates the form Abanki by " terre au
Levant." {Histoire des Abenakis, Introd. p. ii, Quebec, 1866.)
In Cree wapaw, in Chipeway wabi, mean narrows or strait; but
they are derivatives from the root ivab, and mean a light or open
place between two approaching shores, as Chip, wabiganta, or
ivabimagad, " there is a strait between the two shores." (Baraga,
Otchipwe Dictionary.') The name Abnaki is, moreover, no argu-
ment either for or against the eastern origin of the Algonkin stock,
as it was merely a local term applied to a very small branch of it
by the French. Hence M. Petitot's criticisms on the theory under
consideration are misplaced and of no weight.
To what has been said in the text I may add that the Algonkins
who visited Montreal early in the 17th century retained distinct
traditions that they had once possessed the land to the east of that
city, and had been driven south and west by the Huron-Iroquois.
See the Abbe Maurault, Histoire des Abenakis, p. iii, and Wm.
W. Warren, Hist, of the Ojibways, Chap. IV (Minnesota, Hist.
Colls., 1885).
INDEX OF AUTHORS
( The principal references are in full-faced iype.)
Abbott, C. C, 44, 52, 57, 69.
Adair, J-, 61.
Alsop, "G., 14.
Anthony, A., 156, i5i, 219.
Aupaumut, H., 18, 20, 23, 45, 113.
Baraga, J., 35, 59, 62.
Barton, B. S., 146.
Beach, W. W., 115, 125.
Beatty, C, 23, 47, 69, 138.
Bozman, J., 15, 23, 29.
Brainerd, D., 46, 62, 65, 127, 137.
Brickell, J., 64.
Brunner, D. F., 52, 57.
Campanius, T., 66, *75, 96, 116,1 26,
131-
Clark, W. P., 152.
Copway, G., 61, 160, 219.
Cummings, A., 87.
Cuoq, F. H., 71, 105.
Darlington, W., 50.
Darwin, C, 140.
De Laet, 31, 44.
Dencke, C. F., 84.
Denny, E., 86, 94.
Donkers, J., 132.
Drake, S. G., 163.
Duponceau, P. S., 77, 102, 121, 155.
Durant, M., 122.
Eager, 36.
Ettwein, J., 14, 18,47,51,88,132,
229, etc.
Evelin, R., 41.
Fast, C, 125.
Fleet, H., 27.
Force, M. J., 29, 31.
Foulke, W. P., 116.
Gallatin, A., 31, 112, 120.
Gray, A., 149, 155.
Grube, B. A., 83, 256.
Guss, N. L., 14.
Haldeman, S. S., 150, 162.
Hale, H., 12, 17, 18, 36, 95, 112,
156.
Hammond, W. A., 1 10.
Harrison, W. H., 64, 112.
Haven, S. F., 150.
Haywood, J., 17.
Heckewelder, J., 15, 16, 18, 20, 21,
22, 23, 30, 35, 43, 78, 92, 128,
136, 140, 146, 219, etc.
Hendricks, Capt., 21.
Henry, M. J., 37, 45, 86.
Hoffman, W. J., 152,
Holland, F. R., 85.
Hough, 125, 229.
Howse, J., 13, 94, 98, 103, 105.
James, E., 61, 152.
Jogues, I., 255,
Jones, D., 60.
Jones, P., 16.
Johnston, J., 26, 30, 125, 145.
Kalm, P., 46, 50, 52,
Kampman, Rev., 28, 84.
Lacombe, A., 12, 26, 43, 103, etc.
Lawson, J., 61.
Lindstrom, 131.
Long, J., 20.
Loskiel, G. H., 18, 29, 47, 70, 91,
137, 229, etc.
Luckenbach, A., 85.
McCoy, I., 125,
McKenney, T. L., 224.
257
258
INDEX OF AUTHORS.
Mallery, G., 152.
Martin, H., 54.
Maurault, J. A., 256.
Mayer, B., 162.
Meeker, J., 87.
Mezzofanti, Cardinal, 108.
Morgan, L. H., 12, 19, 21, 34, 40,
47. 93-
Morse, J., 31, 113, 145.
Murray, W. V., 24.
Neill, E. D., 27.
Occum, S., 67, 70.
Peale, F., 51.
Peet, S. D., 124.
Penn, Wm., 58, 75, 122.
Petitot, E., 256.
Pickering, J., 94.
Porter, T. C., 57.
Proud, R., 20, 37, 45-
Rafinesque, C. S., 148, etc.
Rasles, S., 60, 94, etc.
Reichel, W. C, 22.
Richardson, J., 58.
Roth, J., 78.
Ruttenber, E. M., 20, 21, 36, 42,
55. "6, 119.
Schmick, J. J., 22.
Schoolcraft, H. R., 20, 58, 62, 87,
109, 133, 160, 219, etc.
Schweinitz, E. de, 25, 62, 129, etc.
Scull, N., 36.
Shea, J. G., 14, 231.
Silliman, B., 155.
Sluyter, Peter, 132.
Smith, G., 38.
Smith, J., 23, 26, 114.
Smith, S., 37.
Squier, E. G., 163, 167, 219, etc.
Stiles, Pres., 35.
Strachey, W., 67.
Tanner, J., 152, 160, 219.
Thomas, C., 17.
Thomas, G., 54, 75, 91, 96.
Thompson, C, 48, 115, 121.
Tobias, G., 87, 88.
Trumbull, J. H., 20, 30, ^;i, 46, 49,
71, 74, 90, 97, 105, 219, etc.
Tryon, G. W., 150.
Van der Donck, 44, 51, 136.
Vincent, F., 60.
Ward, Dr., 153-4.
Wassenaer, 55, 72.
Watson, J.,
Weiser, Conrad, 60, 123.
Whipple, Lt., 87, 96.
White, A., 27, 28.
Wied, Prince of, 55.
Williams, R., 30, 55, 61, 94.
Young, T., 38, 63.
Zeisberger, 35, 55, 62, 69, 76, 105,
113, 129, 134, etc.
INDEX OF SUBJECTS.
( The principal references are in full-faced type.)
Abnaki, ii, 19.
derivation of name, 256.
Age of Gold, 135, 222.
Agozhagauta, 14, 255.
derivation of, 255.
Algonkins, location, 9.
dialects, 11, 89, 93.
dialects, traits of, 89.
myths, 67, 130, 164, 167.
legends, 145.
eastern origin of, 14, 145, 256.
Allemoebi, chief, 123.
Alligewi, 141-2, 229-31.
Alleghany, derivation, 229-3 1.
Alternating consonants, 94.
Andastes, 14.
Arms, native, 53.
Assigunaik, 228.
Assiwikales, 32.
Auquitsaukon, 35.
Bear, Naked, legend of, 146,
Blackfeet, 9, 49, 130.
Bones, preservation of, 25, 54.
Book, Lenape word for, 59.
Brandywine creek, Indians on, 48.
Brant, Joseph, 122.
Brush nets, 53.
Buffalo, the, 226.
Cachnawayes, 26.
Canai. See Conoys.
Canassatego, 15, 114, 121.
Canaways. See Conoys.
Cantico, derivation, 73.
Cape May, tribes at, 41.
Cardinal Points, the, 67.
Carolina, tribes from, 25, 31, 32.
Catawbas, 31.
Cherokees, 13, 16, 166, 230.
Chesapeake Bay, Indians on, 15, 23,
24, 25.
Chicomoztoc, 139.
Chihohockies, 37.
Chiholacki, the, 20, 37.
Chilicothe, 30.
Chipeways, 9, 56, 62, 113, 130, 131,
151-2, 222.
Christina Creek, 15.
Civility, chief, 48.
Cohongorontas, 15.
Condolence, custom of, 18.
Conestoga Creek, 15.
Conestogas, 14.
Confederacy, Algonkin, 19.
Conoys, 25.
Conoy town, 29.
Copper, use of, 50, 52.
Cree dialect, 10, 12, 98.
Crees, 9.
Crosweeksung, or Crosswicks, 45.
Dance, sacred, 73.
Deed, First Indian, 120.
Delamattenos, 16. See Talamatans
and Hurons.
Delawares. See Lenape.
Deluge, Myth of, 134, 167.
Dialects of the Lenni Lenape, 91.
Dogs, 54.
Dreams, belief in, 70.
Dyes, use of, 53.
Eastlanders, 19.
Eries, 13.
Ermomex, 42.
Eskimos, 70, 232.
Fairfield, founding of, 124.
Fire worship, 65, 73.
Fish River, 229.
Five Nations. See Iroquois.
"Four Sticks," the, 152.
Four winds as deities, 65, 67.
259
260
INDEX OF SUBJECTS.
Foxes, tribe, ii, 113.
Friends, their relations to the In-
dians, 63, 126.
Frog Indians, 44.
Ganawese. See Conors.
Gekelemukpechunk, town, 123.
Gesture-speech, native, 152.
Glus-kap, Micmac god, 130.
Gnadenhiitten, 124, 125, 128.
Gollitchy, chief, 118.
Gookin, Governor, 118.
Gordon, Governor, iig.
Grave Creek Mounds, 17.
Grandfathers, Delawares as, 23,
US-
Grandfathers, Fire as, 65, 73.
Guaranis, the, 70.
Hare, the Great, 66.
Head, idols of, 68.
Heart, symbolic meaning of, 71.
Hieroglyphics, native, 57.
Hithquoquean, chief, 117.
Hurons, 13, 16, 144, 165, 168, 231.
Idols, 68.
Indian corn. See Maize.
Indian paths, the, 45.
Inscribed stones, 57.
Interments, 54.
Iroquois, location, 13.
history, no, 114, 120.
Kanawha, derivation, 26.
Kanawhas. See Cottoys.
Kansas, Delawares in, 126.
Kikeron, 132, 133.
Kittawa-Cherokees, 16.
Koquethagachton, chief. See White
Eyes.
Kuscarawocks, 23.
Lenape, the, 33.
myths of, 130.
Lenape dialects, 91, sqq.
prefixes, 99.
grammatical structure, 105.
derivation, 2,Z-
Light, worship of, 65, 130, 132.
Long Island, Indians of, 67, 70.
Long Walk, the, 115, 128.
Machtoga, a festival, 73.
Macocks, 38.
Mahicanni. See Mohegans.
Maize, native name of, 48.
origin of, 228.
Manabozho, 167. See Michabo.
Manito, derivation of, 219.
Mantes, 42, 44.
Manufactures, 51.
Marcus Hook, derivation, 39.
Masco, chief, T45.
Meday worship, 71.
Medicine men, 71, 135.
rattle, 135.
lodge, 71.
Mengwe, derivation, 14, 116, 141.
Mesukkummegokwa, 222.
Miamis, 9, 144, 146.
Michabo, 130, 167.
Micmacs, 10, 48, 130.
Milky Way, myth of, 70.
Mingo, 15, 116, 1 18.
Mingo Creek, 15.
Minisink. See Minsi.
Minquas, 14.
Minsi, 19,36, 114, 116,117, 122,
dialect, 92.
Mission Delaware dialect, 97.
Mohegan dialect, 22, 93.
Mohegans, 19, 20, 165.
myths of, 136, 139.
Monsey. See Minsi.
Montauk Indians, 67.
Mounds, building of, 17, 51,
builders, 231.
Munsees. See Minsi.
Myths of Lenapes, 130.
Namaes sipu, 141, 143.
Nanabozho, 130, 131, 166, 224.
Nanticoke dialect, 24.
Nanticokes, 22, 145.
traditions of, 139.
Narraticons, 42.
Neobagun, the, 151, 152.
Neutral Nation, 13.
New Albion, 41.
New Jersey Lenape, 40, 127, 256.
INDEX OF SUBJECTS.
261
New Jersey Lenape, their dialect,
46, 93. 95-
Ninniwas, 151.
Nottoway s, 13.
Obviative, in Lenape, 107.
Ohio, Delawares in, 124, 125.
Okahokis, 38.
Old Sack, 25.
Olum, derivation of, 153.
Onas, name of Penn, derivation, 95.
Onondagas, 1 17.
Opings, 21, 42.
Opossum, the, 43.
Opuhnarke, the, 19.
Osages, 151, 161.
Ossuaries, 23, 54.
Otayachgo, tribe, 22.
Ottawas, 113, 122, 140, 145, 232.
Paint, word for, 60.
Paints, use of, 53.
Paint Creek, 60.
Palisades, 51.
Pascatoway, derivation, 26.
Pascatoways, 15, 25, 47-
Passive voice, in American lan-
guages, 108.
Peace-belt, the, 47, 114.
Peace chiefs, 47.
Penn, Wm., 75, 116, 122, 127.
his Indian name, 95.
his treaties, 120.
Pequods, 30.
Pictographs, 56.
Pipes, 50, 118.
Piquas, 29.
Piscatoways. See Pascatoways.
Playwickey, derivation, 39.
Pohhegan, the, 35.
Pomptons, 42, 43.
Potomac, Indians near, 25, 67.
Iroquois name of, 15.
Pottawatomies, 11, 113.
Pottery, native, 51.
Powwow, derivation, 70, 227.
Priests, native, 70.
Pueblo Indians, no.
Record Sticks, 59.
Red Score, the, 161.
Sachem, derivation, 46.
Sacs or Sauks, li, 113.
Safe Harbor, inscription, 57.
Sanhicans, 43.
Sapoonies, the, 31.
Scheyichbi, 40, 143.
Scythians, disease of, no.
Senecas, 117, 121.
Serpent worship, 71-2, 167, 222,
231.
Seven, as a sacred number, 139.
Shamokin, 29, 115, 123.
Shawnees, 29, 39, 113, 119, 145,
219.
sacred song of, 145, note.
Shekomeko, 128.
Sign-language, native, 152.
Snake, the Great, 71, 167.
Snake people, the, 165, 227, 231.
land, the, 167, 231.
water, 136.
Soap-stone, use of, 52.
Soul, doctrine of, 69.
Spears, use of, 53.
Stars, knowledge of, 55.
Stockbridge Indians, 45, 113.
Sun worship, 65.
Susquehanna, derivation of, 14.
lands, 120.
Susquehannocks, 13, 53, 116, 121.
Tadirighrones, 31.
Talamatans, 165, 168, 231.
Talega, the, 165-6.
Talligewi, 141-2, 229, 231.
Tamany, 41, 117, 229.
Tatemy, Moses, 128.
Taurus, constellation of, 55.
Tawatawas, 146.
Taway or Tawas, 232.
Tedpachxit, chief, 124-5.
Tedyuscung, 33, 40.
Thahutoolent, chief, 125.
Thousand Isles, the, 165.
Tiawco, the, 22.
Time, computation of, 55.
Tobacco, name and culture, 49, 228.
Tockwhoghs, 23.
Tollan, 225.
Totemic animals, the, 39, 68.
marks, 39, 57.
262
INDEX OF SUBJECTS.
Towanda, derivation, 23.
Tsalaki, 166, 230.
Tula, 225.
Turkey River = Ohio, 39.
Turkey sub-tribe. See Utialacht-
gos.
Turtle, symbol of, 132-35.
Turtle sub-tribe. See Unamis.
Twelve, a sacred number, 73.
Twightees, 146, 232.
Unalachtgo, derivation, 36.
Unalachtgos, 37.
Unami, derivation, 36.
dialect, 79, 80, 91,
Unamis, 37.
Virgin -mother, myth of, 131.
Vowel change in Lenape, 107.
Walam, derivation, 60, 104, 161.
Walam Olum.
evidences of its authenticity,
67, 89, 136, 155-158, 225.
history of, 151.
phonetic system, 159.
metrical form, 159.
Walam Olum.
pictographic system, 160.
MS. of, 162.
synopsis of, 1 64.
Wallamiink, 53, 60.
Wampanos, 21, 128.
Wampum belts, 47, 138.
Wapanachki, the, 19.
Wapeminskink, town, 124.
Wapings, 21, 42, 1 28.
Wappingers, the, 20.
War captains, 47.
Water god, the, 222.
Wendats. See Hurons.
We-shellaqua, 219, 220.
White Eyes, chief, 58, 121, 123.
White River, the, 124, 144, 153.
Winicaco, 24.
Wingenund, chief, 58.
Wiwash, the, 25.
Women, the Lenape as, 109.
Wonameys, 36.
Wolf sub-tribe. See Minsis,
Wyandots, 13, 16, 231.
Year, the native, 55.
Zinzendorf, Count, 128.
LIBRARY
— OF —
ABOIIIGIIillL AMERICim LITERATUIIE,
GENERAL EDITOR AND PUBLISHER:
X). G-. BIE^IIsrTOlsr. -Ml.JD.
The aim of this series of publications is to put within the reach of scholars authentic
materials for the study of the languages and culture of the native races of America.
Each work is the production of the native mind, and is printed in the original
tongue, with a translation and notes, and only such are selected as have some intrinsic
historical or ethnological importance. The volumes of the series are sold separately,
at the prices named.
NOW READY.
No. I. THE CHRONICLES OF THE MAYAS.
Edited by DANIEL G. BRINTON, M.D. 279 pages. Cloth, uncut, $5.00. ($3.00 when a complete
set is ordered.)
This volume contains five brief chronicles in the Maya language of Yucatan, written shortly after
the Conquest, and carrying the history of that people back many centuries. To these is added a
history of the Conquest, written in his native tongue, by a Maya Chief, in 1562. The texts are pre-
ceded by an introduction on the history of the Mayas ; their language, calendar, numeral system, etc. ;
and a vocabulary is added at the close.
No. II. THE IROQUOIS BOOK OF RITES.
Edited by HORATIO HALE. 222 pages. Cloth, uncut, $3.00.
This work contains, in the Mohawk and Onondaga languages, the speeches, songs and rituals with
which a deceased chief was lamented and his successor installed in office. It may be said to throw a
distinct light on the authentic history of Northern America to a period fifty years earlier than the
era of Columbus. The Introduction treats of the ethnology and history of the Huron-Iroquois. A
map, notes and a glossary complete the work.
No. III. THE COMEDY-BALLET OF GUEGUENCE.
Edited by DANIEL G. BRINTON, M.D. 146 pages. Cloth, uncut, $2.50.
A curious and unique specimen of the native comic dances, with dialogues, called daiies, formerly
common in Central America. It is in the mixed Nahuatl-Spanish jargon of Nicaragua, and shows
distinctive features of native authorship. The Introduction treats of the ethnology of Nicaragua,
and the local dialects, musical instruments, and dramatic representations. A map and a number of
illustrations are added.
No. IV. A MIGRATION LEGEND OF THE CREEK INDIANS.
By A. S. GATSCHET. 251 pages. Cloth, uncut, $3.00.
This learned work offers a complete survey of the ethnology of the native tribes of the Gulf States.
The strange myth or legend told to Gov. Oglethorpe, in 1732, by the Creeks, is given in the original,
with an Introduction and Commentary.
No. V. THE LENAPE AND THEIR LEGENDS.
By Dr. DANIEL G. BRINTON. Cloth, uncut, $3.00.
Contains the complete text and symbols, 184 in number, of the Walam Olum or Red Score of the
Delaware Indians, with the full original text, and a new translation, notes and vocabulary. A
lengthy introduction treats of the Lenape or Dela wares, their history, customs, myths, language,
etc., with numerous references to other tribes of the great Algonkin stock.
JJV PREPARATION :
THE ANNALS OF THE CAKCHIQUELS. By Francisco Arana Ernantez
Xahila. With a translation and notes by Dr. D. G. Brinton.
ABORIGINAL AMERICAN ANTHOLOGY. Chiefly original material, fur-
nished by various collaborators.